Evaluating Effects of CRS and Vehicle Features

Evaluating Effects of CRS and Vehicle Features
UMTRI-2010-38
NOVEMBER 2010
EFFECTS OF VEHICLE FEATURES ON
CHILD RESTRAINT INSTALLATION ERRORS
KATHLEEN D. KLINICH, MIRIAM A. MANARY,
CAROL A. C. FLANNAGAN, LAURA J. MALIK,
MATTHEW P. REED
i
Technical Report Documentation Page
1. Report No.
2. Government Accession No.
3. Recipient's Catalog No.
UMTRI-2010-38
4. Title and Subtitle
5. Report Date
Effects of Vehicle Features on CRS Installation Errors
November 2010
6. Performing Organization Code
7. Author(s)
8. Performing Organization Report No.
Klinich, Kathleen D., Manary, Miriam A., Flannagan, Carol A. C.,
Malik, L. J., Reed, M. P.
9. Performing Organization Name and Address
10. Work Unit No. (TRAIS)
University of Michigan Transportation Research Institute
2901 Baxter Rd.
Ann Arbor MI 48109
11. Contract or Grant No.
12. Sponsoring Agency Name and Address
13. Type of Report and Period Covered
VTTI
National Highway Traffic Safety Administration
14. Sponsoring Agency Code
15. Supplementary Notes
Results of Task 1, Labels, Instructions and Features of Convertible Child Restraint Systems (CRS): Evaluating their Effects
on CRS Installation Errors are reported in a separate publication UMTRI-2010-37.
16. Abstract
This report documents a study of how vehicle features contribute to CRS installation errors. Thirty-two subjects were
recruited based on their education level (low or high) and experience with installing CRS (none or experienced). Each
subject was asked to perform four child restraint installations in three vehicles. Each subject first performed a CRS
installation with a seatbelt in one vehicle, followed by three CRS installations using LATCH, one in each of three vehicles.
One child restraint with a hook-on LATCH connector and one with a push-on LATCH connector were used. All
installations were forward-facing, using an 18-month-old CRABI anthropomorphic test device (ATD). Six vehicles were
used in testing, with half of subjects testing with each vehicle. Conditions were selected to provide a range of LATCH
locations (visible, above seating surface, buried in bight), buckle stalk types (webbing vs. rigid), and tether locations
(package shelf vs. seatback). After each installation, the experimenter evaluated 28 factors for each installation (such as
tightness of installation, tether tightness, and LATCH belt attached correctly).
Analyses used linear mixed models to identify the CRS installation outcomes associated with vehicle features. For LATCH
installations, vehicles requiring higher forces to attach connectors to lower anchorages were more likely to be attached
incorrectly. Vehicle seats with a bightline waterfall (which places the lower anchorage above the seating surface) increased
rates of tight CRS installation for both seatbelt and LATCH installs. Seatbelt installations were tight (and locked) more
frequently when the buckle stalk was located close to the bight rather than further forward.
Subjects used the tether correctly in 30% of installations. Subjects used the tether more frequently during LATCH
installations compared to seatbelt installations. The tether was used more frequently in sedans (with anchorage locations on
the package shelf) than in vehicles with the tether anchorage located on the seatback. However, when the tether was used, it
was routed correctly more often in vehicles with the tether anchorage on the seatback. A tether wrap around distance of 210
mm was sufficient to allow tightening of the tether with the two CRS tested, but additional testing showed that 5/16 CRS
could not be tightened sufficiently with this wrap around distance.
Installation time decreased with successive trials, but installation time was longer when subjects used the vehicle or CRS
manuals. Subjects used the vehicle manual in 38% of installations, and were more likely to do so when the tether anchorage
was located on the vehicle seatback. Subjects used the CRS manual in 88% of installations.
In questionnaire responses, subjects indicated that the head restraints affected installations, and vehicle manuals varied in
their ease of understanding. They also noted that tether anchorages on seatbacks were more difficult to locate than those on
the package shelf.
Results from this study do not fully support SAE and ISO recommendations for LATCH usability in vehicles.
Recommendations are made regarding tether anchorage markings, minimum tether wrap around distance, and lap belt
anchorage locations.
17. Key Word
18. Distribution Statement
Child restraints, installation, misuse, ease-of-use,
vehicle features
19. Security Classif. (of this report)
Form DOT F 1700.7 (8-72)
20. Security Classif. (of this page)
Reproduction of completed page authorized
i
21. No. of
Pages
22. Price
Metric Conversion Chart
APPROXIMATE CONVERSIONS TO SI UNITS
WHEN YOU KNOW MULTIPLY
TO FIND
BY
SYMBOL
SYMBOL
In
inches
LENGTH
25.4
Ft
feet
0.305
meters
m
Yd
yards
0.914
meters
m
Mi
miles
1.61
kilometers
km
millimeters
Mm
AREA
squareinches
645.2
square millimeters
mm2
ft2
squarefeet
0.093
square meters
m2
yd2
square yard
0.836
square meters
m2
Ac
acres
0.405
hectares
ha
square miles
2.59
square kilometers
km2
in
2
mi
2
VOLUME
fl oz
fluid 29.57
ounces
milliliters
mL
gal
gallons 3.785
liters
L
3
ft
cubic
feet
0.028
cubic meters
m3
yd3
cubic
yards
0.765
cubic meters
m3
NOTE: volumes greater than 1000 L shall be shown in m3
MASS
oz
ounces
28.35
grams
g
lb
pounds
0.454
kilograms
kg
T
short
tons
(2000
lb)
0.907
megagrams Mg (or "t")
(or "metric
ton")
TEMPERATURE (exact degrees)
o
F
Lbf
Fahrenheit
5 (F-32)/9
or (F-32)/1.8
Celsius
o
C
FORCE and PRESSURE or STRESS
poundforce 4.45
newtons
N
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
lbf/in2
poundforce 6.89
per square
inch
kilopascals kPa
LENGTH
millimeters
0.039
inches in
M
meters
3.28
feet
M
meters
1.09
yards yd
kilometers
0.621
miles mi
Mm
Km
mm2
square
millimeters
AREA
0.0016
ft
square
inches
in2
m2
square meters
10.764
square
feet
ft2
m2
square meters
1.195
square
yards
yd2
Ha
hectares
2.47
acres
ac
square
kilometers
0.386
square
miles
mi2
2
km
mL
milliliters
VOLUME
0.034
fluid ounces
fl oz
liters
0.264
gallons
gal
3
m
cubic meters
35.314
cubic feet
ft3
m3
cubic meters
1.307
cubic yards
yd3
L
MASS
G
grams
0.035
ounces
oz
Kg
kilograms
2.202
pounds
lb
megagrams (or
"metric ton")
1.103
short tons
(2000 lb)
T
Mg (or "t")
TEMPERATURE (exact degrees)
o
C
Celsius
1.8C+32
Fahrenheit
o
F
FORCE and PRESSURE or STRESS
N
Newtons
0.225
poundforce
lbf
poundforce per lbf/in2
square inch
*SI is the symbol for the International System of Units. Appropriate rounding should be
made to comply with Section 4 of ASTM E380.
(Revised March 2003)
kPa
Kilopascals
0.145
iii
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Acknowledgments
This work was funded by the National Highway Traffic Safety Administration under
cooperative agreement DTNH22-09-R-00286 with collaboration between Virginia Tech
Transportation Institute and the University of Michigan.
The authors would like to acknowledge Matthew Jackson, Jamie Moore, and Rachael
Hamilton for their contributions to this project.
iv
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Table of Contents
Metric Conversion Chart..................................................................................................... ii
Acknowledgments.............................................................................................................. iv
Table of Contents ................................................................................................................ v
List of Figures ................................................................................................................... vii
List of Tables ..................................................................................................................... ix
Executive Summary ............................................................................................................ 1
Introduction ......................................................................................................................... 5
Overview ......................................................................................................................... 5
Scope of the Problem ...................................................................................................... 5
The LATCH System and Misuse .................................................................................... 6
Ease of Use Ratings ........................................................................................................ 8
Vehicle Manuals ........................................................................................................... 10
Methods............................................................................................................................. 12
Recruitment and Subject Selection ............................................................................... 12
CRS Selection ............................................................................................................... 12
Vehicle Selection .......................................................................................................... 13
Test Setup...................................................................................................................... 15
Testing Sequence .......................................................................................................... 16
Testing Forms ............................................................................................................... 17
Test Matrix .................................................................................................................... 18
Vehicle Manual Assessment ......................................................................................... 18
Data Analysis ................................................................................................................ 19
Results ............................................................................................................................... 21
Installation Assessment ................................................................................................. 21
Overview ................................................................................................................... 21
CRS installation tightness ......................................................................................... 22
Lower anchorages ..................................................................................................... 26
Tether ........................................................................................................................ 27
Seatbelt installations ................................................................................................. 31
Manual use ................................................................................................................ 32
Securement factors .................................................................................................... 34
Subject installation time............................................................................................ 35
Subject Questionnaires.................................................................................................. 36
Subject ratings of installations .................................................................................. 36
Subject ratings of vehicles and methods ................................................................... 41
Subject identification of CRS installation methods .................................................. 42
Evaluation after last install........................................................................................ 46
Discussion ......................................................................................................................... 50
Summary and comment on results ................................................................................ 50
Installation tightness ................................................................................................. 50
Lower anchorage use ................................................................................................ 51
Tether use .................................................................................................................. 51
Seatbelt installations ................................................................................................. 54
Instruction manual use .............................................................................................. 54
Education/experience ................................................................................................ 55
v
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Subject assessments .................................................................................................. 55
Subject understanding of CRS installation options .................................................. 56
Post-test evaluation of subject feedback ................................................................... 57
Comparison to Task 1 results.................................................................................... 57
Limitations .................................................................................................................... 58
Recommendations ......................................................................................................... 60
References ......................................................................................................................... 61
Appendix A ....................................................................................................................... 62
Appendix B: ...................................................................................................................... 65
Appendix C ....................................................................................................................... 70
Appendix D ....................................................................................................................... 77
vi
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
List of Figures
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.
Figure 9.
Figure 10.
Figure 11.
Figure 12.
Figure 13.
Figure 14.
Figure 15.
Figure 16.
Figure 17.
Figure 18.
Figure 19.
Figure 20.
Figure 21.
Figure 22.
Figure 23.
Figure 24.
SAE tool for checking anchorage collinearity. ............................................... 9
SAE tool for measuring force required to attach lower connector. .............. 10
SAE tool for checking clearance around lower anchorages. ........................ 10
SAE tool to determine if any tether routing device provides sufficient
clearance for tether hook and adjustment hardware...................................... 10
Recaro Signo (left) [C6] and Evenflo Triumph (right) [C10] CRS used in
Task 2 testing. ............................................................................................... 13
Tool to measure force required to engage lower anchorage (top view with
force gauge on left and side view of connector portion on right). ................ 14
Percentage of installations passing 1” tightness test by subject education and
experience. .................................................................................................... 22
Percentages of installations passing 1” tightness test for LATCH and seatbelt
installations by lower anchorage location. .................................................... 23
Illustration of vehicle seats with a bightline waterfall with vertical lower
anchorage locations indicated. ...................................................................... 23
Percentages of installations passing 1” tightness test for LATCH and seatbelt
installations by lower anchorage force.......................................................... 24
Percentages of installations passing 1” tightness test for LATCH and seatbelt
installations by lower anchorage location depth within the bight. ................ 24
Percentages of installations passing 1” tightness test for LATCH and seatbelt
installations by seatbelt buckle location. ...................................................... 25
Percentages of installations passing 1” tightness test for LATCH and seatbelt
installations by CRS model. .......................................................................... 25
Side views of C10 (left) and C6 (right). ....................................................... 26
Percentage of LATCH installs with correct lower attachment by lower
anchorage force ............................................................................................. 26
Percentage of installations where tether is used correctly by subject
education and experience. ............................................................................. 27
Percentage of tether installs where tether is tight by subject education and
CRS experience. ............................................................................................ 28
Percentage of installations according to tether use by installation method and
tether anchorage location. ............................................................................. 29
Percentage of installations where tether is attached correctly by fore-aft
location of base of head restraint. ................................................................. 29
Percentage of installations using tether where tether is tight by fore-aft
location of base of head restraint. ................................................................. 30
Percentage of tether installations where tether is tight vs. tether wrap around
distance. ........................................................................................................ 31
Percentage of seatbelt installations where the seatbelt is locked by subject
experience and education level. .................................................................... 31
Percentage of seatbelt installations where the seatbelt is locked by buckle
location. ......................................................................................................... 32
Percentage of installations in which subject used vehicle manual by subject
education and experience. ............................................................................. 33
vii
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 25. Percentage of installations in which subject used CRS manual by subject
education and experience. ............................................................................. 34
Figure 26. Percentage of installations where harness is snug by subject experience and
education level. ............................................................................................. 34
Figure 27. Percentage of installations where harness is snug vs. percentage of tight
installations for each vehicle. ........................................................................ 35
Figure 28. Average installation time by trial. ................................................................. 36
Figure 29. Subject assessments that did not vary with vehicle type. ............................. 37
Figure 30. Mean subject rating of whether vehicle manual matched CRS manual by
vehicle (each color represents a manufacturer)............................................. 37
Figure 31. Mean subject rating of whether head restraint made it hard to install by
vehicle (each color represents a manufacturer)............................................. 38
Figure 32. Subject responses to questions regarding all installations. ........................... 39
Figure 33. Subject responses to questions regarding LATCH installations. .................. 39
Figure 34. Subject responses to questions regarding seatbelt installations. ................... 40
Figure 35. Mean subject responses about understanding vehicle manual by vehicle. ... 40
Figure 36. Subject responses to ease of finding tether anchorage by vehicle and tether
location. ......................................................................................................... 41
Figure 37. Average subject ratings of vehicles for LATCH and seatbelt installations. . 42
Figure 38. Diagram of options for installing CRS in each test vehicle. S: seatbelt,
L:LATCH, T: tether, *:Center LATCH installation with inboard LA from
outboard positions ......................................................................................... 43
Figure 39. Distribution of subject responses regarding available methods for CRS
installation in different vehicle seating positions. ......................................... 44
Figure 40. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by trial. ............................................... 45
Figure 41. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by number of diagrams in the CRS
portion of the vehicle manual........................................................................ 45
Figure 42. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by vehicle. .......................................... 46
Figure 43. Post-test instruction placard. ......................................................................... 47
Figure 44. Subject responses to instruction placard on tightening. ................................ 48
Figure 45. Subject responses to instruction placard on harness snugness. ..................... 49
Figure 46. Representative tether anchorage under marked door in sedans. ................... 52
Figure 47. Representative unmarked tether anchorage often found in minivans/crossover
vehicles. ........................................................................................................ 52
Figure 48. Tether anchorage located too close to allow space to tighten tether. ............ 53
Figure 49. Incorrectly routed but tight tether installation. ............................................. 54
viii
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
List of Tables
Table 1.
Table 2.
Table 3.
Table 4.
Table 5.
Table 6.
Table 7.
Table 8.
Summary of vehicle and subject factors that affect installation ..................... 2
Percentage of correct installations: average Task 1A results for four CRS .. 13
Characteristics of vehicles selected for study ............................................... 15
Test matrix * ................................................................................................. 19
Factors considered as predictors in analysis ................................................. 20
Summary of CRS and subject factors that affect installation ....................... 21
Vehicle head restraints with average subject rating of difficult (hardest to
easiest) ........................................................................................................... 56
Comparison of misuse rates (percentages) in Task 1A and Task 2 testing .. 58
ix
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Executive Summary
This document reports on a study that evaluated the effect of vehicle features on child
restraint system (CRS) installation errors and installation experiences. This Task 2 effort
complements the work of Task 1, which assessed the effect of CRS features, labels, and
instructions on CRS installation errors.
Thirty-two volunteers were recruited based on general education (either “have not
attended college” or “graduated from college”) and CRS installation experience levels
(none or experienced). People who had attended a child car seat check or who had been
trained as a NHTSA CPS technician were barred from participation. Each volunteer was
assigned to perform four forward-facing installations: one using a seatbelt and one using
LATCH in one vehicle, as well as LATCH installations in two other vehicles. The
installations were performed in the right second-row seating position (five vehicles) or
right third-row seating position (one vehicle).
Two child restraints used in Task 1 were selected for testing. One has a push-on style of
LATCH connector (Recaro Signo), while the other has a hook-on LATCH connector
(Evenflo Triumph). The harness of each CRS was configured to fit the 18MO CRABI
dummy used in testing, so subjects could focus on attaching the CRS to the vehicles
rather than restraining the dummy.
Six vehicles were selected for testing, with each subject performing installations in three
of the vehicles. The vehicles were selected to provide a variety of lower anchorage
locations (visible or buried, and at or above seating surface), forces required to attach the
lower connectors (high, medium, or low), seatbelt buckle locations (at or forward of the
bight), head restraint prominence (more forward or more rearward), and tether locations
(package shelf or seatback).
Following each installation, the experimenter assessed the installation, checking for key
installation factors such as tightness, correct attachment of tether and LATCH belt, and
locking the seatbelt. Subjects also answered questions regarding the ease of the
installation. Table 1 summarizes different installation factors and the predictor variables
associated with each.
1
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Table 1. Summary of vehicle and subject factors that affect installation
Percentage Predictors
F-test
p-value
Correct
Installation
67%
LA location (above seating F(1,30)=15.12
0.0005
tightness
surface > bight)
(1” test)
Buckle location (seatbelt
F(1,30)=5.65
0.0241
only; bight> forward)
CRS x installation method F(1, 94.44)=5.04 0.0271
(C6>C10 for LATCH,
C10>C6 for belt)
Lower
90%
LA attachment force (low, F(2,55)=5.84
0.0050
anchorages
med > high)
attached
LA visibility (visible >
F(1,56)=4.38
0.0409
correctly
hidden)
Tether attached
30%
TA location (SB>PS)
F(1,80)=10.27
0.0019
correctly
Installation method
F(1,80)=21.86
<0.0001
(LATCH > seatbelt)
HR fore-aft location
F(4,80)=3.26
0.0157
(RW>FW)
Tether tight
82%
Installation method
F(1,92.73)=29.66 <0.0001
(LATCH > seatbelt)
HR fore-aft location
F(4,75.03)=2.15 0.0833
(FW>RW)
Tether location (SB>PS)
F(1,39.1)=4.87
0.0332
Seatbelt locked
50%
Buckle location (at
F(1,29.38)=8.65 0.0063
bight>forward)
Installation time NA
Trial (1>2>3>4)
F(1,94.7)=44.77 <0.0001
Use of vehicle manual
F(1,22.1)=20.31 0.0002
(use>nonuse)
Use vehicle
38%
TA location (SB>PS)
F(1,31)=10.58
0.0028
manual
Use CRS manual 88%
CRS experience
F(1,32.13)=5.94 0.0205
(inexperienced >
experienced)
* LA = lower anchorage; TA = tether anchorage; HR = head restraint; SB=seatback;
PS=package shelf; FW=forward; RW=rearward; C6=Recaro Signo; C10=Evenflo
Triumph.
Visible lower anchorages (LA) were associated with higher rates of correct attachment of
the LATCH belt compared to lower anchorages that were buried within the bight, but LA
visibility did not affect the rate of achieving a tight CRS installation. The same is true of
the force required to attach the lower connector, with vehicles requiring lower/medium
forces having more correct LATCH belt attachments than those requiring higher
securement forces.
2
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Some vehicles had a bightline waterfall feature (a bolster cushion located directly behind
the seated pelvis location) that places the lower anchorages above the seating surface.
This higher location was associated with a higher rate of tight installation for both
LATCH and seatbelt installations, suggesting that the shape of the seat may help
installers to obtain a tight installation. This finding does not support the SAE CRS
committee recommendation to keep lower anchorages close to the seating surface as a
means of improving LATCH usability.
For seatbelt installations, buckle locations closer to the bight resulted in tight installations
more often than those located forward of the bight (83% vs. 35%). However, location of
seatbelt buckle did not have an effect on LATCH installs, indicating that the seatbelt
buckle locations tested did not interfere with use of lower anchorages.
The tether was used correctly in only 30% of all installations, and not used in 26% of all
installations. Subjects used the tether in 85% of LATCH installations and 48% of
seatbelt installations. When used, it was used correctly in 45% of LATCH installations
and 30% of seatbelt installations. The most common error in tether use was routing the
tether strap over an adjustable head restraint, instead of routing underneath or removing
the head restraint as directed by the vehicle manual prior to installation.
The location of the tether anchorage had a major effect on tether use. The tether was
more likely to be used in both LATCH and seatbelt installations in sedan-type
configurations where the tether anchorage is located on the package shelf. Tether use
was lower when the anchorage was located on the vehicle seatback, where it is usually
found in minivans and crossover vehicles. Subjects overwhelmingly indicated that the
seatback location for the tether anchorages was more difficult to find in these three
vehicles. As a result, subjects were twice as likely to use the vehicle manual in these
vehicles compared to sedans. These findings support requirements for uniform marking
of tether anchorages in vehicles.
The tether anchorage zone is defined in FMVSS 225 relative to an R-point determined
using the H-point machine (SAE J826). FMVSS 225 also defines a wrap around distance
that describes the path length between the tether anchorage and an estimated top of the
child restraint. A problem reported from the field is that some tether anchorages in the
most forward portions of this zone do not permit enough space for the tether hook and
attachment hardware to allow adequate tightening of the tether. In the vehicle with a
tether wrap around distance of 210 mm, all of the tether installations were tight. In the
vehicle with the tether wrap around distance of 180 mm, less than 40% of the
installations were tight, and tight installations could only be achieved by incorrectly
routing the tether over the head restraint. Additional evaluations with 14 additional CRS
indicated that 5 of 16 CRS cannot be tightened with a tether wrap around distance of 210
mm. Additional research is needed to determine a minimum tether wrap around distance
in vehicles that is compatible with tether hardware and attachment location on CRS.
Subjects used the vehicle manual in 38% of installations and consulted the CRS manual
in 88% of installations. Subjects gave different ratings for each vehicle as to whether the
vehicle manual agreed with the CRS manual. Subjects also gave a variety of scores to
3
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
each vehicle with regard to understanding the CRS installation instructions included in
each vehicle manual. The two vehicle owner manuals with the shortest CRS sections,
which also had the fewest diagrams, had the highest scores, while the manual with the
longest CRS section (and the most diagrams) received the lowest rating on ease of
comprehension.
The only vehicle feature noted in the subject questionnaires as making it harder to install
the CRS was the head restraint. The two questions having the widest range of responses
were the ease of finding the tether anchorage and understanding the vehicle manual.
After performing one seatbelt installation and three LATCH installations, a preference for
LATCH installation was expressed by 80% of subjects at the end of their session.
Unlike Task 1, no subjects withdrew from the study, and only one did not finish all four
trials. In Task 2, dummy installation was much less complicated, because the harness
height was already adjusted to fit the dummy. This seemed to reduce subjects’ frustration
levels, making them less likely to give up once they reached the CRS installation portion
of the trial. In addition, installations were all forward-facing. In Task 1, although
performance did not generally vary between forward-facing and rear-facing installations,
rear-facing installations averaged twice as long to install. Finally, the two CRS selected
for the current study had the highest rates of correct installation in Task 1.
The results of both Task 1 and Task 2 of this study indicate that the vehicle and CRS
features recommended by experts in child passenger safety may not lead to correct
performance by regular CRS users. This finding suggests that testing with non-expert
volunteers should be performed to validate any recommendations by experts intended to
improve ease of child restraint installation.
4
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Introduction
Overview
This document reports on the second portion of a two-part study examining the effects of
child restraint system (CRS) features, labels, and manuals and vehicle features on CRS
installation errors and installation experiences. Results of Task 1 are reported separately
(Klinich et al., 2010). The first phase of Task 1 compared volunteer installation
experiences using 16 different CRS installed either forward-facing or rear-facing in a
single vehicle. The second phase of Task 1 tested volunteer installation experiences
using different modified labels and manuals for two CRS. In Task 2, documented in the
current report, the focus was on the influence of vehicle features on seatbelt and LATCH
installations.
Scope of the Problem
In the U.S., motor-vehicle crashes are the leading cause of death for children ages 3-18
(CDC 2007). In 2008, 1633 children under the age of 16 died and 220,000 were injured
as a result of motor vehicle crashes (NHTSA, 2009 Early Edition of the 2009 Traffic
Safety Facts Annual Report). The use of a child restraint system (CRS) is an effective
countermeasure that reduces the likelihood of a child crash fatality by 71% for infants
and 54% for toddlers, depending primarily on the restraint type and orientation (NHTSA,
2002). Misuse has been shown to markedly reduce the effectiveness of CRS (Carlsson et
al., 1991; Decina et al., 1994; Johnston at al., 1994; Ruta et al., 1993; Bilston et al.,
2007). Several studies have estimated CRS misuse rates ranging from 73 - 94% (Eby and
Kostyniuk, 1999; Decina and Lococo 2005; Koppel and Charlton, 2009; Lane et al.,
2000). Some of the variation in these estimates originates in the study designs, subject
recruitment methods, and the level of inspection that is used to determine misuse.
Identified types of misuse observed in the field include:












Loose vehicle seatbelt
Loose harness straps
Incorrect selection of CRS for height/weight/age of child
Improper positioning of harness strap
Improper harness belt routing
Improper vehicle belt path
Unbuckled vehicle seatbelt
Harness not used
Harness not buckled
CRS broken or damaged
Less than 80% of the CRS base footprint located on/above the vehicle seat
Inappropriate CRS installation angle
5
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report








Incorrect CRS direction (i.e. using an infant seat forward-facing)
Nonuse of a tether, when available and appropriate
Incorrect tether strap tensioning
Use of both LATCH (Lower Anchors and Tethers for Children) and seatbelt to
secure a CRS
Placement of a rear-facing (RF) CRS in front of an active frontal airbag.
Improper harness retainer clip position
Improper retainer clip threading
Attachment of aftermarket products to the restraint
Loose CRS-to-vehicle installation and loose occupant restraint harness have been
consistently observed across studies as the most frequent types of misuse. Lane (2000)
surveyed the CRS installations for 109 subjects and found that 84% had between one and
three installation errors, with an average of two errors per installation.
Several studies have identified factors correlated with misuse. Koppel and Charlton
(2009) found statistically significant differences in misuse rates between CRS types, with
forward-facing (FF) harness restraints having a higher observed level of misuse than rearfacing seats or belt positioning boosters. Eby and Kostyniuk (1999) found that higher
levels of misuse were associated with: lower educational levels, situations where the
driver was not the child’s legal guardian, the number of times that the seat was
moved/reinstalled into different vehicles, and with children who were younger and
smaller. Lane et al. (2000) found a trend for less misuse with higher education
attainment level and participation in a private insurance program.
The LATCH System and Misuse
The Lower Anchors and Tethers for Children (LATCH) system, consisting of two lower
anchors and an anchorage for a tether attached to the top of the child restraint, was
phased into the US market beginning in September 1999. Federal Motor Vehicle Safety
Standard (FMVSS) 225, Child Restraint Anchorage Systems, specifies requirements for
tether and lower anchorage system hardware to be installed in vehicles. The standard,
which establishes requirements for the locations and strength of the anchorage systems,
was first established in 1999 and was revised most recently in 2004. Two of the main
reasons to implement LATCH were (1) to provide an easier method for CRS installation
that would eliminate the need to know how to lock the seat belt system or use a locking
clip and (2) to increase the use of top tethers to reduce forward head excursion, and in
turn, head contacts during crash events.
In 2007, Decina and Lococo published the results of a misuse survey focusing
specifically on LATCH. Their findings show that in situations where tether use was
required and all the tether hardware was available, only 51% of those surveyed were
using the top tether. Loose tethers were observed in 18% of cases and loose LATCH
straps were seen in 30% of cases. In 20% of cases, CRS were installed using both
6
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
LATCH and seatbelt. This study highlighted that the availability of LATCH did not
eliminate CRS misuse.
LATCH hardware has been implemented in the vehicle fleet such that it meets the
requirements of FMVSS 225, which are summarized in Appendix A. However, a number
of problems have been reported, including:
 Unclear labeling of top tether anchors
 Presence of other hardware (such as cargo tie downs) with a similar
appearance that are mistakenly used as tether anchorages
 Some combinations of lower anchorages and CRS attachment hardware that
are incompatible and prevent securement of the CRS using LATCH,
 Lower anchorage locations that are buried so deeply in the bight that they are
inaccessible,
 Interference with use of an adjacent vehicle seating position when a CRS is
installed with LATCH.
Issues regarding the shape of the vehicle seat and head restraint and how they interact
with a CRS installed with LATCH have also been identified. A CRS installed with
flexible LATCH may tip laterally an unacceptable amount because of the shape of the
vehicle seat contour. Vehicle head restraint design changes stemming from the recent
revision of FMVSS 202 (Head Restraints) may lead to interference between the head
restraint and an installed CRS or the child restraint fixture (CRF). The test procedure for
FMVSS 225 does not dictate the position of the head restraints when using the CRF or a
maximum permissible force for CRF installation.
Child passenger safety advocates indicate a need for additional LATCH anchorages in
vehicles, particularly in center seating positions (Stewart et al, 2009). In most cases, the
width of vehicle seats prevents accommodation of three pairs of lower anchorages spaced
280 mm apart. Instead, some, but not all, vehicle manufacturers indicate that the inboard
lower anchorages from two outboard seating positions can be used to secure a CRS in the
center seating position. Installing additional lower anchorages in a seating row may
block access to the lower anchorages or vehicle belts in adjacent seating positions.
The location requirements for tether and lower anchorages have led to some CRS
installation problems in the field. In some cases, there has been a discrepancy between
the length of tethers provided with the CRS and the allowable tether anchorage zone that
may result in tethers that are too short to reach the tether anchorages. Some tether
anchorages subject the tether hook to bending loading or require twisting of the tether to
attach the hook. No procedure or requirement is available to ensure sufficient space
between the lower anchorages and the CRS to allow adjustment of flexible attachment
length.
Since FMVSS 225 was enacted, ISO has developed standardized symbols for marking
lower and tether anchorages that might be beneficial to include in the U.S. standard. The
requirements for positioning labels for lower anchorages in FMVSS 225 is focused on
visibility of the symbols while the ISO labeling requires the symbol to be within a
7
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
specified distance of the hardware location. In some cases a symbol that is close enough
to the lower anchorages to meet the ISO requirement is not visible per FMVSS 225.
Although FMVSS 225 indicates that tether and lower anchorages must be accessible at
all times, the level of accessibility varies. Some vehicles have tether anchorages that
cannot be accessed when the CRS is installed on the vehicle seat. The wide variety of
tether anchor locations in the cargo areas of SUVs and minivans can make them difficult
to find, and lack of uniform labeling requirements make them difficult to distinguish from
other vehicle structures such as cargo tie-down points. In some cases, the “tunnel”
devices intended for improving lower anchorage accessibility may hinder access,
especially when releasing the LATCH attachments.
Ease of Use Ratings
The current NHTSA Ease of Use (EOU) Rating system (NHTSA, 2006) was developed
to identify CRS with features that enhance usability and to provide consumers with this
information. NHTSA does not currently address vehicle features in its ease-of-use
program.
In the field, some misuse modes arise from features of the vehicle environment and
others result from interactions between specific CRS and vehicle combinations. A
usability rating scheme under development in the ISO Child Restraints Group has rating
forms for all three situations: the CRS, the vehicle, and specific combinations of the two
(ISO, 2008). This rating system currently focuses on LATCH-type systems that are
called ISOFIX systems in the international arena. Some of the vehicle features that are
rated in the current version of the ISO document include the vehicle manual instructions
on how to identify the number and location of seating positions available for CRS
installation, the visibility and labeling of the LATCH anchors, the presence of other
hardware elements that could be mistaken for LATCH anchors, the actions required for
preparing the seating position for CRS installation, and conflicts between LATCH and
seatbelts. Adding a vehicle ease-of-use rating to the current NHTSA evaluation could
present another opportunity for reducing CRS installation errors.
Procedures and tools for assessing LATCH usability and compatibility with CRS have
been drafted by the SAE Child Restraint Systems Subcommittee (SAE, 2009). Although
the procedures have been evaluated by some subcommittee members, no widespread
testing of the procedures has been conducted. Thus the procedures have not been
validated to determine if their recommendations result in more usable LATCH
anchorages that reduce errors in CRS installation. The SAE CRS committee has
developed the following recommendations to improve LATCH usability: (SAE CRS
Committee 2009)
1) The distance between tether anchorage and CRS must be large enough to allow
tightening of the tether.
2) Tether routing devices must be sufficiently large to accommodate all types of tether
hook and adjustment hardware.
3) Lower and tether anchorages should be visible or clearly marked by standard
symbols.
8
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
4) The seat cushion contour should allow most of the base of the CRS to contact the
seat cushion and permit placement of the CRS at reasonable lateral and pitch angle.
5) Components around the lower anchorages should not interfere with attachment of
lower connectors.
6) The stiffness of the seat components near the lower connectors should not require
excessive force to attach the lower connectors.
7) Lower anchorages should be close to the seating surface to permit good contact
between the vehicle seat and the bottom of the CRS.
8) Visible lower anchorages are considered the easiest to use, because hidden anchors
increase likelihood of incorrectly attaching to other hardware.
9) Using LATCH in a particular seating position should not affect use of seatbelts or
LATCH in adjacent seating positions.
10) To allow use of rigid LATCH CRS connectors, each pair of lower anchorages
must meet a collinearity specification.
A summary of the SAE tools and procedures that are included in the draft SAE
recommended practice for assessing these points is included below.
1) At each seating position, completely engage the SAE child restraint fixture (CRF)
specified in FMVSS 225 with the lower anchorages.
2) Install the CRF using rigid connectors. Measure the distance between the seat cushion
and lower reference point on the CRF (maximum recommendation of 5 cm [2 in]).
Measure the pitch angle of the CRF (recommendation of 15° +5°/-10°).
3) Measure the gap between the installed CRF base and the seat cushion at a point
400 mm forward of the rear reference point on the CRF. (Currently no
recommendation.)
4) Place the CRF at each LATCH seating position with the CRF rigid connectors
retracted. Measure the lateral angle of the CRF seat (recommendation of +/-5°).
5) Attach collinearity gauge shown in Figure 1 to ensure that the anchorages have been
manufactured within an acceptable tolerance.
Figure 1. SAE tool for checking anchorage collinearity.
6) Measure the force required to attach the SAE tool shown in Figure 2 to each lower
anchorage (recommendation less than 15 lb).
9
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 2. SAE tool for measuring force required to attach lower connector.
7) When 15 lb of force is applied to the tool shown in Figure 3 while it is attached to the
lower anchorages, measure the angle relative to horizontal (recommendation is 75
degrees above horizontal).
Figure 3. SAE tool for checking clearance around lower anchorages.
8) When attached to lower anchorage, rotate gauge shown in Figure 3 to make sure it
does not contact any rigid structure (which would demonstrate potential for releasing a
push-button LATCH connector).
9) Ensure that any tether routing guides are large enough to accommodate the SAE gauge
shown in Figure 4.
Figure 4.
SAE tool to determine if any tether routing device provides sufficient
clearance for tether hook and adjustment hardware.
Vehicle Manuals
An issue repeatedly identified as leading to misuse of CRS is the difficulty of
understanding instructions and labels. A review of the literature on this topic is found in
the companion report on Task 1 (Klinich et al., 2010). However, most research has
10
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
focused on the labels and instructions provided with the CRS, and no researchers have
studied the role of vehicle manuals on CRS installation.
As described earlier in this introduction, FMVSS 225 specifies requirements for vehicle
manuals regarding use of the LATCH system. However, no other federal regulations
specify requirements regarding CRS installation using the seatbelt. Information provided
by vehicle manufacturers is voluntary, and the content varies widely.
One attempt to provide clear consumer information about CRS installation using LATCH
is The LATCH Manual published by Safe Ride News (Stewart et al., 2009). It is a
comprehensive reference guide published every other year that provides basic
information for Child Passenger Safety technicians and consumers who want to use the
LATCH system to secure a child restraint in their vehicle. The manual covers general
CPS background information with a focus on LATCH and top tether best practices. It
also provides a make and model guide to LATCH-related vehicle features and their
recommended usage. One of the strengths of the publication is that it provides specific
LATCH information that is difficult to obtain and is not always supplied in the vehicle or
CRS owner’s manual. For example, the LATCH manual gives specific part numbers for
vehicle retrofit tether anchor kits and explicitly states the manufacturers’ mass limit for
use of LATCH to secure CRS.
11
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Methods
Recruitment and Subject Selection
The recruitment process and target subject pool for Task 2 were the same as those used in
Task 1. Subjects were primarily recruited through Internet advertisements on Craigslist
and AnnArbor.com, the local online newspaper, as well as through online mailing lists of
local elementary and high schools.
Thirty-two volunteers were recruited based on general education and CRS experience
levels. Subjects were classified in a lower education level if they had not gone to college,
and in the higher education level if they had graduated from college. Subjects were also
classified by their child restraint installation experience as either “none” or
“experienced.” To be considered experienced, the subject had to have installed at least
two different types of CRS in two different vehicles multiple times in the last five years.
People who had completed the NHTSA CPS course or had attended a child seat check up
event were disqualified from participation. The same subject recruiting script and
consent form used in Task 1 (Klinich et al., 2010) were used in Task 2.
All testing conducted for this project was approved by a University of Michigan
institutional review board (IRB) that reviews protocols of research programs involving
human subjects.
CRS Selection
One CRS with each of the two common types of LATCH connectors (push-on and hookon) was selected from the pool of 16 different CRS models used in Task 1 of the study.
Also, because the focus of Task 2 is on the vehicle factors, two CRS that were relatively
uncomplicated to install were selected.
Table 2 shows the average rate of correct installation among trials for four of the bestperforming CRS from Task 1A. These two CRS with push-on connectors and two CRS
with hook-on connectors had the highest average rate of correct installation among all
these factors. These data suggested that the best candidate CRS for this study were the
Signo and Triumph because they had higher average scores on installation tasks
compared to the Boulevard and Scenera. Trial installations with these CRS in the
proposed test vehicles indicated that they could be installed well with either LATCH or
seatbelt, so they were selected for use in the remaining testing. To maintain consistent
coding with Task 1, the Signo is labeled C6 and the Triumph C10. Pictures of these two
CRS are shown in Figure 5.
12
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Table 2.
Signo
Boulevard
Triumph
Scenera
Percentage of correct installations: average Task 1A results for four CRS
Connector
type
Pushbutton
Pushbutton
Hook-on
Hook-on
1" test
43%
Belt
routing
100%
Angle
100%
LATCH
correct
100%
Harness
slot
29%
Harness
snug
57%
Mean
63%
75%
88%
83%
50%
63%
70%
29%
29%
100%
100%
86%
71%
86%
71%
57%
43%
86%
43%
74%
60%
72%
Figure 5. Recaro Signo (left) [C6] and Evenflo Triumph (right) [C10] CRS used in
Task 2 testing.
Vehicle Selection
Six vehicles were selected for testing. Ideally, for each factor under consideration, two
vehicles should have a similar relevant feature. Each subject performed four installations
in three of the test vehicles, which allowed subjects to perform two installs within a
single vehicle (one seatbelt and one LATCH).
The factors considered for vehicle selection were prioritized based on input from
NHTSA, recommended SAE practices for improving LATCH compatibility in vehicles,
factors considered in the ISO vehicle assessment rating, and the researchers’ experience
as CPS technicians. When choosing the vehicles for testing, the goal was to find six
vehicles with the following range of features, listed by highest priority:
1)
2)
3)
4)
5)
Different lower anchorage locations, such as above seating surface, visible,
buried
Different seat stiffness around the LATCH anchorage
Different seatbelt buckle characteristics, such as rigid and webbing and
location relative to the bight
Protruding and non-protruding head restraints
Variable tether locations
The ability to rent a vehicle for the applicable test period while staying within the project
budget was also considered. Other vehicle factors that may contribute to CRS installation
13
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
were documented, but not necessarily used as a main factor in vehicle selection,
including seat contour, seat stiffness, different manufacturers, covers over the lower
anchorages, leather vs. fabric seats, number of available seating positions, method of
locking seatbelt, cushion angle, and seatback angle.
Thirty vehicles were evaluated for consideration in the study. One measurement in the
vehicles was the force required to engage the SAE tool shown in Figure 6 with the lower
anchorages. The tool combines a generic LATCH connector shape with a digital force
gauge. The SAE recommended practice for LATCH usability suggests that the measured
force should be no more than 15 lb.
Figure 6. Tool to measure force required to engage lower anchorage (top view with
force gauge on left and side view of connector portion on right).
Seatbelt factors include whether the buckle stalk is flexible or rigid, and the location of
the belt anchor relative to the bight. The type of buckle stalk can potentially affect access
to the lower anchorages in LATCH installations, and also affect the ability to route the
seatbelt and get a tight seatbelt installation. Based on the researchers’ experience, buckle
locations forward of the bight can sometimes be more difficult to use with CRS
installation than those located at the bight, because the belt geometry does not allow the
CRS to be pulled back towards the vehicle seatback.
Table 2 shows the list of vehicles selected for this study, as well as a description of some
key characteristics. Data for the vehicle used in Task 1 (2006 Pontiac G6) are also
included and labeled vehicle G. The data shown are for the seating position the subjects
were directed to use, which was the second row right for all vehicles except the Sienna,
which was the third row right. Appendix B shows detailed photos of each vehicle used in
testing. The locations of the lower anchorages and seat belts relative to the bight were
qualitative assessments. For the measured forces, low corresponds to attachment forces
of 2-13 lbf, medium 14-18 lbf, and high corresponds to 26-29 lbf.
14
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Vehicle
Ford Flex
Honda
Civic
Toyota
Sienna
Ford
Fusion
Dodge
Avenger
Chrysler
Pacifica
Pontiac
G6
Table 3. Characteristics of vehicles selected for study
Code Lower
Lower
Head
Seat
Anchorage
Anchorage restraint
belt
Force
Protrude?
A
Buried at bight
High
No
Rigid,
Forward
B
Buried, bight
Low
No
Webbing,
above seating
Forward
surface
C
Visible at bight
Low
Yes
Rigid,
Bight
D
Buried at bight
Medium
Yes
Webbing,
Forward
E
Buried, bight
High
Yes
Webbing,
above seating
Bight
surface
F
Visible at bight
Medium
No
Rigid,
Bight
G
Visible at bight
Medium
No
Webbing,
Forward
Tether
Location
Back of
seat
Package
Shelf
Back of
seat
Package
Shelf
Package
Shelf
Back of
seat
Package
shelf
Subjects performed installations in either vehicle group ABC or DEF. Each of these
groups exposes the subject to three lower anchorage locations. For each pair of vehicles
with the same lower anchorage position, the vehicles have different lower anchorage
forces (high/medium, low/high, low/medium). The vehicles were selected so half had
protruding head restraints and the other half did not, based qualitative assessment of
protrusion. In addition, half have tether locations on the vehicle seatback and half were
on the rear package shelf. For the seatbelt, half of the vehicles have the anchorage at the
bight, with the other half forward of the bight. At least one vehicle in each group had a
rigid seatbelt stalk, while the rest were anchored with flexible webbing.
For the purposes of designing the matrix, tether locations were classified as either on the
package shelf or on the back of the seatback. The actual locations and implementations
of the tether anchorages vary within these categories. While the main comparisons in
analysis were between the two main categories of tether anchorages, possible effects of
different tether locations/implementations were also considered. In a similar manner, the
Civic and Avenger both have the bights located above the seating surface and are in one
LATCH location category even though the heights above the seat differ. The potential
effect of this was considered in analysis.
Test Setup
For each vehicle, the experimenter prepared the vehicle environment for testing by
adjusting the rear head restraints to the lowest position and adjusting the front seats to
their mid-track fore/aft location and adjusting to the seatback angle to the seatback
recline one notch back from upright, pre-test position. The indoor testing area was
15
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
arranged with the three vehicles in a semicircle so the subject could approach all three
test vehicles from the right side. Both rear-side doors and the right-front door were left
open throughout the session. In the one vehicle where testing was performed in the third
row, the second-row seats were removed or folded flat.
Still cameras were used to record side and isometric views of each CRS installation.
Exceptional or unusual installations were also photographed. A digital video camera was
located on a tripod near the driver door to record installations in the right-rear seat. A
wireless microphone placed in the vehicle enhanced sound recording. During subject
recruitment, subjects were asked if they agreed to be videotaped, and none of them
declined. The subjects were encouraged but not required to “think aloud” during the
installation. The videos were used to check notable installations, but complete analysis
of the videos was beyond the scope of the project.
The 18MO CRABI crash test dummy (ATD) was used for testing. The dummy was
dressed in a sweatshirt and sweatpants and was placed in the CRS by the subjects for
each installation. In Task 1, many subjects experienced frustration when trying to adjust
and secure the ATD in the harness. To focus subjects’ effort on installing the CRS rather
than securing the ATD in Task 2, the CRS harness was initially configured to
accommodate the 18MO ATD, meaning the harness slot height was adjusted to the
position closest to the shoulders, and the crotch strap was pre-positioned to fit the
dummy. This approach is different from Task 1, where the subject had to configure the
CRS to fit the ATD. Because the CRS was configured to fit the dummy, the only two
child securement factors that were assessed were the harness snugness and whether it was
buckled properly.
Each CRS was marked with the subject ID number, the trial number, the CRS code, the
test vehicle code and the date for photographs. Optional padding and cupholders were
removed from the CRS. However, the CRS was otherwise configured in its “out-of-thebox” state for the recline position, tether position, and LATCH belt position. The
original labels and manuals for each CRS were used.
The experimenter prepared a test cart that contained potentially helpful materials
available to the subject during the test session. The cart held the CRS instruction
manuals, the test dummy and a flathead screwdriver. The vehicle manuals were stowed
in their normal storage location in the vehicle. The subjects were told they could use the
vehicle and CRS manuals, and the experimenter was allowed to help the subject find the
vehicle manuals if needed and identify parts on the CRS.
Testing Sequence
Appendix C contains the script that was used during testing. If asked questions, the
experimenter told the subject that the information could be found in the CRS and vehicle
manuals. The experimenter helped the subject to find the vehicle manual if help was
requested.
16
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Because Task 1 did not identify any factors predicting whether subjects chose to use
LATCH or seatbelt to install the CRS, each subject was assigned a method of installation
for each trial. They were asked to install the CRS with the seatbelt in the first trial and to
use LATCH for the remaining three trials. If the subject could not perform an installation
with the directed method, they were asked to try to use the other method instead. For all
test vehicles but the Sienna, subjects were asked to install the CRS in the right secondrow seating position, because the vehicle was selected for the study based on the
characteristics of the LATCH system in that seating position. For the Sienna, the subjects
were directed to install the CRS in the right third-row position.
Because Task 1 did not find any difference in correct installation rates between forwardfacing and rear-facing on key tasks such as obtaining a tight installation and routing the
belt, the subjects were asked to perform all installations forward-facing. Choosing to use
forward-facing installations for the test conditions meant that all trials should involve
tether use. Focusing on one direction of installation also allowed more statistical power
for assessing the vehicle factors.
Appendix C also contains the forms used to evaluate the CRS installation, which were
revised slightly from Task 1 to clarify some items regarding how the subject locked the
seatbelt. In addition, factors regarding child securement (harness slot position, chest clip
height) were not documented. The methods of measuring CRS tightness and tether slack
used in Task 1 were also used in Task 2. Definitions of each measure are also included in
the Appendix.
Testing Forms
Appendix D contains post-test evaluation forms that were filled out by the subject. The
first part of the subject form asks the subjects whether they agree or disagree with
statements about installing the CRS. The second part asks them to rate the ease of
different parts of the installation. The forms are modified slightly from those used in
Task 1, partly to address more vehicle factors and partly to clarify points where subjects
were confused in Task 1. For example, separate forms were prepared for LATCH and
seatbelt installations, because subjects were often confused during Task 1 when using
previous questionnaires (i.e., responding to questions about lower anchorages for seatbelt
installations). If the subject had questions about terminology when filling in the form,
such as “Is this the tether?” the experimenter identified the item for the subject. The
subject filled out the form behind a screen so they could not see the experimenter
assessing the CRS installation, but was allowed to come and look at the vehicle, CRS,
labels, or instructions if they wanted to do so.
To gauge the subjects’ ability to identify LATCH hardware in the vehicle, the final part
of the assessment asks the subject to indicate on a diagram of vehicle seating positions
which ones can be used to install LATCH, which ones can be used for seatbelt
installation, and which ones have a tether anchorage.
17
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
After the last installation, subjects were asked whether they found LATCH or seatbelt to
be easier to use. In addition, they were asked to rate each vehicle installation on a scale
of one to ten.
Test Matrix
Table 4 shows the test matrix for Task 2, based on a split plot design. Because more of
the key vehicle selection factors are related to LATCH, the matrix is weighted towards
more LATCH installations than seatbelt installations. Each subject was asked to install
the CRS using the seatbelt in the first vehicle, then perform LATCH installations in the
other two vehicles for the second and third installations, and perform a LATCH
installation in the first vehicle for the fourth installation. Each subject performed two
installations with each child restraint.
Vehicle Manual Assessment
The child restraint installation section of each vehicle manual was reviewed to generate
potential predictor variables for data analysis. The readability of the CRS section of each
manual was calculated using readability tools in Microsoft Word. The number of pages
dedicated to CRS installation, as well as the number of diagrams in the CRS section,
were also counted to describe each section.
18
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Table 4. Test matrix *
Subject
Installation order
First Second Third Fourth
X0L
1
AS2 BL2
CL1
AL1
2
BS2 CL2
AL1
BL1
3
CS1 AL2
BL1
CL2
4
AS1 BL2
CL1
AL2
5
DS2 EL2
FL1
DL1
6
ES2 FL2
DL1
EL1
7
FS1 DL2
EL1
FL2
8
DS1 EL2
FL1
DL2
X2L
1
CS2 AL2
BL1
CL1
2
AS2 BL2
CL1
AL1
3
BS1 CL2
AL1
BL2
4
CS1 AL2
BL1
CL2
5
DS2 EL2
FL1
DL1
6
ES2 FL2
DL1
EL1
7
FS1 DL2
EL1
FL2
8
DS1 EL2
FL1
DL2
X0H
1
BS2 CL2
AL1
BL1
2
CS2 AL2
BL1
CL1
3
AS1 BL2
CL1
AL2
4
BS1 CL2
AL1
BL2
5
FS2 DL2
EL1
FL1
6
DS2 EL2
FL1
DL1
7
ES1 FL2
DL1
EL2
8
FS1 DL2
EL1
FL2
X2H
1
AS2 BL2
CL1
AL1
2
BS2 CL2
AL1
BL1
3
CS1 AL2
BL1
CL2
4
AS1 BL2
CL1
AL2
5
ES2 FL2
DL1
EL1
6
FS2 DL2
EL1
FL1
7
DS1 EL2
FL1
DL2
8
ES1 FL2
DL1
EL2
*ABCDEF=vehicles; L,S= LATCH/Seatbelt; 1,2=CRS
Data Analysis
Table 5 lists the variables that were considered as potential predictors in analysis. The
analysis methods used in Task 1 were also used in Task 2. Prior to performing the main
statistical analysis using linear mixed models and generalized linear mixed models, each
dependent measure was reviewed using univariate analysis to identify the most relevant
potential predictors for each dependent variable. In addition to the types of misuse
19
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
documented for each installation, dependent variables included the responses from the
subject assessment forms completed for each test session.
Table 5. Factors considered as predictors in analysis
Type of factor
Variable
Subject
Education
CRS Experience
Gender
Vehicle
Head restraint protrusion
Seat contour
Seat stiffness
Distance from estimated H-point to base of head restraint
Angle from estimated H-point to base of head restraint
Test Conditions
CRS
Trial
Attachment method (seatbelt/LATCH/both)
Lower Anchorage Visible/buried
At or above seating surface
Depth into bight
Attachment force (low, medium, high)
Tether
Overall location (package shelf or back of seat)
Approximate wrap distance
Design style (under door, visible, hidden)
Seatbelt
Buckle stalk type (webbing/rigid)
Buckle location (at or forward of bight)
Locking method (retractor or latchplate)
Vehicle Manual
Grade Level/Reading Ease
# diagrams in CRS section
# pages in CRS section
20
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Results
Installation Assessment
Overview
Table 6 summarizes different installation factors and the predictor variables associated
with each. More details about each factor are found in the following sections.
Table 6. Summary of CRS and subject factors that affect installation
Percentage Predictors
F-test
p-value
Correct
Installation
67%
LA location (above seating F(1,30)=15.12
0.0005
tightness
surface > bight)
(1” test)
Buckle location (seatbelt
F(1,30)=5.65
0.0241
only; bight> forward)
CRS x installation method F(1, 94.44)=5.04 0.0271
(C6>C10 for LATCH,
C10>C6 for belt)
Lower
90%
LA attachment force (low, F(2,55)=5.84
0.0050
anchorages
med > high)
attached
LA visibility (visible >
F(1,56)=4.38
0.0409
correctly
hidden)
Tether attached
30%
TA location (SB>PS)
F(1,80)=10.27
0.0019
correctly
Installation method
F(1,80)=21.86
<0.0001
(LATCH > seatbelt)
HR fore-aft location
F(4,80)=3.26
0.0157
(RW>FW)
Tether tight
82%
Installation method
F(1,92.73)=29.66 <0.0001
(LATCH > seatbelt)
HR fore-aft location
F(4,75.03)=2.15 0.0833
(FW>RW)
Tether location (SB>PS)
F(1,39.1)=4.87
0.0332
Seatbelt locked
50%
Buckle location (at
F(1,29.38)=8.65 0.0063
bight>forward)
Installation time NA
Trial (1>2>3>4)
F(1,94.7)=44.77 <0.0001
Use of vehicle manual
F(1,22.1)=20.31 0.0002
(use>nonuse)
Use vehicle
38%
TA location (SB>PS)
F(1,31)=10.58
0.0028
manual
Use CRS manual 88%
CRS experience
(inexperienced >
experienced)
F(1,32.13)=5.94
0.0205
* LA = lower anchorage; TA = tether anchorage; HR = head restraint; SB=seatback; PS=package shelf;
FW=forward; RW=rearward; C6=Recaro Signo; C10=Evenflo Triumph.
21
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
CRS installation tightness
Percentage of installations passing 1"
test for tightness
The percentage of installations passing the 1” tightness test is shown in Figure 7 for each
education/experience group. Neither subject CRS experience level nor subject education
level had a statistically significant affect on installation tightness.
100%
90%
75.6%
80%
72.7%
66.7%
70%
60%
50%
46.2%
40%
30%
20%
10%
0%
Low Edu
Hi Edu
Inexperienced
Low Edu
Hi Edu
Experienced
Figure 7. Percentage of installations passing 1” tightness test by subject education
and experience.
The rate of tight installation is shown separately for LATCH and seatbelt installations in
the subsequent graphs as a function of different vehicle features. Figure 8 shows the rates
for vehicles where the lower anchorage location is above the seating surface compared to
vehicles where the lower anchorage is at the seat bight located at the intersection of the
vehicle seat cushion and seatback. Lower anchorage location was a significant predictor
of installation tightness for both LATCH and seatbelt installations [F(1,30)=15.17;
p=0.0005]. Since lower anchorage location would not be expected to be associated with
the tightness of seatbelt installations, this suggests that the shape of vehicle seats which
result in the lower anchorage being positioned above the seating surface is likely
responsible for the higher rates of tight installation. The two vehicles with the lower
anchorages above the seating surface have a bightline waterfall feature as shown in
Figure 9.
22
Percentage of installations passing 1:"
test for tightness
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
100%
90%
LATCH
86%
82%
Seatbelt
80%
70%
62%
60%
50%
38%
40%
30%
20%
10%
0%
Above
Bight
Lower Anchorage Location
Figure 8. Percentages of installations passing 1” tightness test for LATCH and
seatbelt installations by lower anchorage location.
Figure 9. Illustration of vehicle seats with a bightline waterfall with vertical lower
anchorage locations indicated.
The percentage of tight installations in relation to the lower anchorage force (see Table 3)
required to attach the lower connectors is illustrated in Figure 10. The difference in tight
installation rates did not vary significantly with the amount of force required to attach the
lower connectors. This is expected for seatbelt installations (which should have no
dependence on force required to attach the LATCH belt), but is somewhat unexpected for
the LATCH installations, as the SAE recommended practice recommends a force of 15 lb
or less to attach the LATCH connectors, which corresponds to the low force category.
23
Percentage of installations passing 1:"
test for tightness
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
80%
70%
LATCH
64%
60%
73%
71%
Seatbelt
56%
54%
46%
50%
40%
30%
20%
10%
0%
Low
Medium
High
Lower Anchorage Force
Figure 10. Percentages of installations passing 1” tightness test for LATCH and
seatbelt installations by lower anchorage force.
Percentage of installations passing 1:"
test for tightness
The rates of correct installation according to the depth of the lower anchorage placement
within the vehicle seat bight are shown in Figure 11. Vehicles with the deepest lower
anchorage locations (4-6 cm) had slightly higher rates of tight installation compared to
those that were visible (0-2 cm) or slightly buried within the bight (2-4 cm). However,
these differences were not statistically significant once other factors are included in the
model.
100%
90%
LATCH
80%
Seatbelt
70%
60%
83%
64%
63%
55%
54%
45%
50%
40%
30%
20%
10%
0%
Low
Med
High
Depth of LA
Figure 11. Percentages of installations passing 1” tightness test for LATCH and
seatbelt installations by lower anchorage location depth within the bight.
The type of buckle stalk (webbing or rigid) did not affect installation tightness. However,
the rate of tight installation according to the location of the seatbelt buckle (either at or
forward of the bight) is shown in Figure 12. Tight installation rates were higher for
seatbelt installation when the buckle is located at the bight, but were similar for LATCH
24
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of installations passing 1:"
test for tightness
installation [F(1,30)=5.65; p=0.0241]. This suggests that seatbelt locations do not
interfere with LATCH installations, but that tight seatbelt installations are more likely
when the belt buckle is closer to the bight.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
LATCH
83%
73%
Seatbelt
67%
35%
Bight
Forward
Seatbelt Buckle Location
Figure 12. Percentages of installations passing 1” tightness test for LATCH and
seatbelt installations by seatbelt buckle location.
Percentage of installations passing
1:" test for tightness
Rates of tight installation varied with CRS and the installation method [F(1, 94.44)=5.04;
p=0.0271]. As shown in Figure 13, CRS C10 had similar rates of tight installation for
LATCH and seatbelt installations, while CRS C6 had a higher rate of tight installation
with LATCH compared to seatbelt. Side views of each CRS are shown in Figure 14.
C10 has the same belt path for both the seatbelt and LATCH. C6 has different belt paths,
with the path for the vehicle belt higher than the LATCH belt path.
90%
LATCH
80%
70%
60%
80%
Seatbelt
60%
60%
50%
40%
40%
30%
20%
10%
0%
C10 (Same belt path)
C6 (different belt paths)
CRS
Figure 13. Percentages of installations passing 1” tightness test for LATCH and
seatbelt installations by CRS model.
25
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 14. Side views of C10 (left) and C6 (right).
Lower anchorages
Overall, the lower connectors were attached correctly to the lower anchorages in 90% of
LATCH installations. Correct attachment is defined as appropriately attaching the CRS
hardware to the correct vehicle hardware in the correct orientation. As shown in Figure
15, there are slight differences in correct use of the LATCH belt associated with lower
anchorage force. Rates of correct LATCH belt attachment were higher when the force
required to attach the anchorages was in the low or medium range compared to the high
range [F(2,55)=5.84; p=0.005]. In addition, when the lower anchorages were visible, the
LATCH belt was correctly attached in all cases, whereas it was attached correctly in 88%
of installations where the lower anchorage was buried within the bight [F(1,56)=4.38;
p=0.0409].
Percentage of LATCH installations
with correct lower attachment
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
96%
100%
79%
Low
Medium
High
Lower anchorage force
Figure 15. Percentage of LATCH installs with correct lower attachment by lower
anchorage force
26
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
The rate of correctly attaching the CRS to the lower anchorages also varied slightly with
the model of the CRS, although results were not statistically different. C10, which has a
hook-type LATCH belt connector, was correctly used 89% of the time. In comparison,
C6, which has a push-on LATCH belt connector, was attached correctly 95% of the time.
Tether
Tether use was considered correct if the subject used the correct anchorage, attached the
hook correctly, and routed the tether relative to the head restraint as directed by the
vehicle and child restraint manuals. The tightness of the tether was evaluated separately
because attachment and tightening can be considered separate installation steps.
Percentage of installations wher tether
used correctly
Overall, 30% of installations used the tether correctly, 44% used it incorrectly, and 26%
did not use the tether at all. The percentage of installations where the tether is used
correctly is shown in Figure 16 for each education/experience group. Although the
higher education, CRS-experienced subject group had the highest rates of correct tether
use, subject education and experience were not significant predictors.
100%
90%
80%
70%
60%
50%
40%
39.4%
30.8%
30%
26.8%
24.2%
Hi Edu
Low Edu
20%
10%
0%
Low Edu
Inexperienced
Hi Edu
Experienced
Figure 16. Percentage of installations where tether is used correctly by subject
education and experience.
Figure 17 shows the percentage of installations where the tether was tight, given that the
tether was used, for each subject education/experience group. The inexperienced subjects
were slightly more likely to tighten the tether appropriately than the subjects with CRS
experience.
27
Percentage of installations with tether
where tether is tight
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
97.0%
100%
90%
85.0%
80%
70.0%
69.6%
Low Edu
Hi Edu
70%
60%
50%
40%
30%
20%
10%
0%
Low Edu
Hi Edu
Inexperienced
Experienced
Figure 17. Percentage of tether installs where tether is tight by subject education and
CRS experience.
Tether use according to the installation method and location of the tether anchorage are
shown in Figure 18. This plot shows the percentage of installations where the tether was
used correctly, the tether was used incorrectly, and the tether was not used. Overall,
subjects were more likely to use the tether with LATCH installations compared to
seatbelt installations [F(1,80)=21.86; p=<0.0001]. They were also more likely to use the
tether in sedans (where the tether anchorage is located on the package shelf and easier to
see from the point of installation) than in minivans/crossover vehicles (where the tether
anchorage is located on the seatback). However, they were more likely to use the tether
correctly when the tether anchorage location was on the seatback rather than the package
shelf [F(1,80)=10.27; p=0.0019]. The most common error in tether use was routing the
tether over the top of the head restraint, rather than routing it underneath or removing the
head restraint as directed by the vehicle user manual. When used, tethers were more
likely to be tight in LATCH installations compared to seatbelt installations [83% vs.
73%, F(1,92.73)=29.66, p<0.0001]. (The denominator for this percentage excludes
conditions where the tether was not used).
28
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of installations
100%
90%
Tether Not Used
80%
Tether Incorrect
70%
Tether Correct
60%
50%
40%
30%
20%
10%
0%
Package shelf
Seatback
Package shelf
LATCH
Seatback
Seatbelt
Tether anchorage
location
------------Installation method
Figure 18. Percentage of installations according to tether use by installation method
and tether anchorage location.
Percentage of installations with tether
attached correctly
The vehicles tested were categorized qualitatively according to whether the head restraint
was more forward or more rearward. Vehicles with the base of the head restraint more
rearward had a higher rate of correct tether use than those when the base of the head
restraint was more forward, in both LATCH and seatbelt installations [F(4,80)=3.26;
p=0.0157]. However, as shown in Figure 20, the rate of tight installation when the tether
was used was marginally higher for the vehicles when the base of the head restraint was
more forward [F(4,75.03)=2.15, p=0.0833].
100%
90%
LATCH
80%
Seatbelt
70%
60%
60%
50%
40%
30%
20%
18%
16%
8%
10%
0%
Rearward
Forward
Base of head restraint
Figure 19. Percentage of installations where tether is attached correctly by fore-aft
location of base of head restraint.
29
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of tether installations
where tether is tight
100%
93%
90%
80%
70%
LATCH
72%
83%
Seatbelt
63%
60%
50%
40%
30%
20%
10%
0%
Rearward
Forward
Base of head restraint
Figure 20. Percentage of installations using tether where tether is tight by fore-aft
location of base of head restraint.
When the tether was used, it was tight in 76% of installations when the tether anchorage
was located on the package shelf, and in 86% of installations when the tether anchorage
was located on the seatback [F(1, 39.1)=4.87, p=0.0332]. Examination of the percentage
of tether installations where the tether is tight vs. the tether wrap around distance as
shown in Figure 21 helps explain why this is the case. The tether wrap around distance
was measured relative to an approximate R-point, with the tether routed relative to the
head restraint as directed by the vehicle manual. The lowest rates are found for the
vehicle with a wrap around distance of 185 mm. The 100% rate of tight tether
installation for the vehicle with a wrap around distance of 210 mm indicates that the
threshold for the minimum required distance for achieving a tight tether installation may
lie somewhere between these two values. As a check to determine how much the
minimum wrap around distance depends on the CRS used, the fourteen other CRS used
in Task 1 were installed using LATCH in the vehicle with the wrap around distance of
210 mm. Overall, 11 out of 16 CRS tethers were able to be tightened with this wrap
around distance.
30
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of tether installations
where tether is tight
100%
90%
80%
70%
60%
50%
40%
30%
LATCH
20%
Seatbelt
10%
0%
0
200
400
600
800
1000
Tether wrap around distance (mm)
Figure 21. Percentage of tether installations where tether is tight vs. tether wrap
around distance.
Seatbelt installations
Overall, subjects locked the seatbelt in 49% of seatbelt installations, either by using the
lockoffs on the CRS, switching the retractor to locking mode, or using a locking
latchplate. Figure 22 shows the percentage of seatbelt installations where the subjects
locked the seatbelt by subject experience and education levels. Education level made a
difference with experienced subjects but not inexperienced subjects.
Percentage of seatbelt installations
where seatbelt is locked
100%
90%
80%
66.7%
70%
60%
54.5%
50.0%
50%
38.9%
40%
30%
20%
10%
0%
Low Edu
Hi Edu
Inexperienced
Low Edu
Hi Edu
Experienced
Figure 22. Percentage of seatbelt installations where the seatbelt is locked by subject
experience and education level.
31
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of seatbelt installations
where seatbelt is locked
As shown in Figure 23, subjects more often locked the seatbelt in conditions where the
buckle was located near the seat bight rather than more forward on the seat bench
[F(1,29.38)=8.65; p=0.0063)]. In addition, rates of locking the seatbelt were slightly
higher with CRS C6 than C10 (53% vs. 45%), probably because C6 is equipped with belt
lockoffs. However, these rates were not statistically different.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
75%
35%
At Bight
Forward of Bight
Buckle location
Figure 23. Percentage of seatbelt installations where the seatbelt is locked by buckle
location.
One vehicle had a locking latchplate (where little or no action is required by the installer
to lock the seatbelt) and the belt was locked 100% of the time when seatbelt installations
were performed in this vehicle (n=4). When the vehicle had a switchable retractor,
seatbelts were locked in 42% of seatbelt installations, likely because this type of locking
mechanism requires the installer to take steps to change the seatbelt into the locked usage
mode. Because of the small number of cases and the strong differences from seatbelt
location, the differences in locking mechanism are not statistically significant.
Another factor assessed was whether the subjects locked the seatbelt as they were
directed to do so by the instructions, which occurred in 13 out of 15 cases. For the two
cases where the subject used an alternate locking method to the one suggested, one
subject used the locking clip, and in the other case, the subject did not use the lockoffs
provided on the CRS.
In designing the study, half of the selected vehicles had webbing-mounted buckles, while
the other half had rigid buckle stalks. The type of buckle stalk did not affect the quality
of the CRS installations.
Manual use
Subjects consulted the vehicle manual in 38% of installations. Subjects used the vehicle
manual but not the CRS manual in only two trials. The percentage of installations where
the subject used the vehicle manual is shown in Figure 24 according to subject experience
32
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
and education level. Experienced, low education subjects were less likely to use the
vehicle manual than subjects in the other categories, but there were no significant
differences with subject education or CRS experience.
Percentage of installations where
subject used vehicle manual
100%
90%
80%
70%
60%
50%
46.2%
41.5%
39.4%
40%
24.2%
30%
20%
10%
0%
Low Edu
Hi Edu
Inexperienced
Low Edu
Hi Edu
Experienced
Figure 24. Percentage of installations in which subject used vehicle manual by
subject education and experience.
Only one vehicle factor had a significant effect on whether subjects used the vehicle
manual. When the tether anchorage was located on the package shelf, 25% of subjects
consulted the vehicle manual. In contrast, when the tether anchorage was located on the
seatback, 52% of subjects used the vehicle manual [F(1,31)=10.58; p=0.0028].
Overall, subjects consulted the child restraint manual in 88% of installations. Figure 25
shows the percentages of installations where the subject used the CRS manual by subject
education and experience. Experienced subjects, particularly those with lower education,
were less likely to use the child restraint manual [F(1,32.13)=5.94; p=0.0205]. No other
vehicle factors made it more likely for subjects to use the child restraint manual.
33
Percentage of installations where
subject used CRS manual
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
96.2%
100%
90%
95.1%
87.9%
80%
70%
60%
50%
40%
30%
20%
72.7%
10%
0%
Low Edu
Hi Edu
Low Edu
Inexperienced
Hi Edu
Experienced
Figure 25. Percentage of installations in which subject used CRS manual by subject
education and experience.
Securement factors
The CRS harness was correctly buckled around the dummy in 91% of trials. No potential
predictors were significantly associated with this performance measure.
Subjects were able to achieve a snug harness in 70% of installations. The distribution of
snug harnesses by subject education and experience is shown in Figure 26. The
differences between subject groups were not statistically significant.
Percentage of installations where
harness is snug
100%
87.2%
90%
75.8%
80%
70%
60%
66.7%
58.3%
50%
40%
30%
20%
10%
0%
Low Edu
Hi Edu
Inexperienced
Low Edu
Hi Edu
Experienced
Figure 26. Percentage of installations where harness is snug by subject experience
and education level.
34
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
No vehicle factors predicted the rate of harness snugness. However, as shown in Figure
27, a slight correlation was observed between the rate of tight installation and the rate of
snug harness [F(1, 121.5)= 13.82; p=0.0003]. This trend could be associated with the
skills and effort of the subjects, but it could also suggest that the subjects may try harder
to correctly secure the dummy if they have not become overly frustrated during the CRS
installation. In this phase of testing, the adjustments needed to fit the harness to the
ATD, such as routing the harness through the correct slots in the CRS shell, were
properly adjusted before the installation trial began. This eliminated additional steps
needed by the subjects and likely contributed to the shorter installation times and higher
incidence of correctly snugged harness.
100%
Percentage of snug harness
90%
80%
70%
R² = 0.48
60%
50%
40%
30%
20%
10%
0%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Percentage of tight installations
Figure 27. Percentage of installations where harness is snug vs. percentage of tight
installations for each vehicle.
Subject installation time
In this phase of testing, none of the subjects withdrew from the study, and only one
subject did not complete all four installations. The average time of installation is shown
in Figure 28, showing that subjects improved their installation time with each trial
[F(1,94.7)=44.77; p<0.0001].
35
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Average installation time (min)
30
25
25
20
20
16
15
10
10
5
0
1
2
3
4
Trial
Figure 28. Average installation time by trial.
Average installation times were shorter for LATCH (16 minutes) compared to seatbelt
installations (23 minutes), but this likely results from most subjects performing seatbelt
installations first. Use of vehicle manuals also affected mean installation times. Subjects
who used the vehicle manual averaged installation times of 23 minutes vs. 15 minutes for
those who did not [F(1,22.1)=20.31, p=0.0002]. Subjects who used the CRS manual
averaged 19 minutes, vs. 9 minutes for those who did not, but this difference is not
statistically significant once effects of trial and vehicle manual use are considered.
(Subjects used the vehicle manual but not the CRS manual in only two trials).
The experimenter’s qualitative assessment of subject performance was compared to
testing in Task 1. Subjects in Task 2 exhibited less frustration than those in Task 1,
probably because all of the installations were performed forward-facing and the harness
height was already adjusted to fit the dummy.
Subject Questionnaires
Subject ratings of installations
Subjects filled out an evaluation form following each CRS installation. Mean values of
subject assessments were calculated for each vehicle, and ANOVA tests were performed
to determine if there were significant differences in subject assessments between
vehicles. Mean results for the first set of questions are shown in Figure 29 for the
questions where the mean values did not vary with vehicle type. Subjects did not think
that the vehicle seat shape, the vehicle seat stiffness, or the seatbelt buckles caused a
problem during installation. They also believed that they attached the child seat to the
vehicle correctly, and what they did during testing is similar to what they would do at
home.
36
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Do you agree with these statements?
The seatbelt buckles got in the way of using
LATCH.
2.1
The shape of the vehicle seat made it hard to
install.
2.4
The stiffness of the vehicle seat made it hard to
install.
2.2
What I did today is similar to what I would do at
home to install a child seat.
3.9
4.1
I attached the child seat to the vehicle correctly.
1
2
Strongly disagree
3
Neutral
4
5
Strongly Agree
Figure 29. Subject assessments that did not vary with vehicle type.
As shown in Figure 30, when the subject asked whether the vehicle manual matched the
CRS manual, results were marginally different among vehicles (p=0.054). Different
colors represent different manufacturers. Mean results ranged from neutral to positive.
Do you agree: Vehicle manual matched CRS manual
3.6
F
3.6
Vehicle Code
E
3.0
D
3.2
C
3.8
B
4.1
A
1
2
Strongly disagree
3
Neutral
Disagree
4
Agree
5
Strongly Agree
Figure 30. Mean subject rating of whether vehicle manual matched CRS manual by
vehicle (each color represents a manufacturer).
Subjects also assessed whether they thought the head restraint on the vehicle seat made it
hard to install the CRS, as shown in Figure 31 (p=0.006). Vehicle E, which subjects
37
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
considered to be the most challenging in terms of the interaction with the head restraint
during CRS installation, has a fixed head restraint.
Do you agree: Head restraint made it hard to install
2.1
F
3.4
Vehicle Code
E
3.0
D
2.4
C
2.9
B
2.4
A
1
2
Strongly disagree
3
Neutral
Disagree
4
Agree
5
Strongly Agree
Figure 31. Mean subject rating of whether head restraint made it hard to install by
vehicle (each color represents a manufacturer).
For most of the remaining subject questions, there were no differences in mean values by
vehicle tested. Figure 32 shows mean subject responses for questions related to all
installations, Figure 33 shows mean subject responses for questions relating to LATCH
installations, and Figure 34 shows mean subject responses for questions relating to
seatbelt installations. All mean values for these subject responses ranged from neutral to
easy.
38
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
How hard or easy was it to:
3.4
Store the top tether (if not used)
Tighten the tether strap on the top of the child
seat
Attach the tether strap on the top of the child
seat to the vehicle
3.4
3.6
3.5
Adjust the angle of the child seat
3.4
Figure out what angle the child seat should be
Understand the instruction manual about
installing the child seat
3.3
3.4
Understand the labels on the child seat
1
2
Very hard Hard
3
Neutral
4
Easy
5
Very easy
Figure 32. Subject responses to questions regarding all installations.
How hard or easy was it to:
3.8
Tighten the LATCH belt
3.7
Attach the LATCH belt connectors to the lower
anchorages
4.0
Find the lower anchorages in the vehicle
3.8
Figure out where to route the LATCH belt
1
2
Very hard Hard
3
Neutral
4
Easy
5
Very easy
Figure 33. Subject responses to questions regarding LATCH installations.
39
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
How hard or easy was it to:
3.6
Store the LATCH belt
Use the lock-offs on the child seat that pinch the
vehicle belt
3.1
3.7
Figure out how to lock the seat belt
3.6
Tighten the vehicle seat belt
3.6
Figure out where to route the vehicle belt
1
Very hard
2
Hard
3
Neutral
4
Easy
5
Very easy
Figure 34. Subject responses to questions regarding seatbelt installations.
Two subject questionnaire responses varied with vehicle type. Figure 35 shows mean
subject responses to ease of understanding the vehicle manual regarding the child
restraint installation (p=0.066). Each color corresponds to a vehicle manufacturer. The
manual with the worst rating (Vehicle C) had the highest number of diagrams in the CRS
section (n=47). Vehicles E and F, which had two of the best ratings, had the least number
of pages in the vehicle manual regarding CRS installation (11 pages), and only 2 or 5
diagrams in the CRS installation section.
How hard or easy was it to understand vehicle manual
about installing the child seat?
3.9
Vehicle Code
F
3.5
E
3.3
D
2.9
C
3.5
B
3.4
A
1
2
Very hard
Hard
3
Neutral
4
Easy
5
Very Easy
Figure 35. Mean subject responses about understanding vehicle manual by vehicle.
40
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Subject assessments of ease of finding the tether anchorage is shown in Figure 36, with
different colors corresponding to vehicles with tether locations on the package shelf or
vehicle seatback. Subjects gave significantly lower ratings to tether anchorages located
on the seatback compared to those located on package shelves (p<0.0001). For the
vehicles tested, the tether anchorages located on the package shelves were also more
clearly labeled than those on the vehicle seatbacks.
How hard or easy was it to find the tether anchorage?
3.1
F
4.4
Vehicle Code
E
3.9
D
2.6
C
Tether location:
4.0 Package shelf
B
Seatback
2.8
A
1
2
Very hard
Hard
3
Neutral
4
Easy
5
Very Easy
Figure 36. Subject responses to ease of finding tether anchorage by vehicle and tether
location.
Subject ratings of vehicles and methods
When asked to choose which installation method they preferred, 79% of subjects chose
the LATCH method. When giving each installation a rating (scale of 1-10, 10 best) on
how much they liked the vehicle based on child restraint installation, the mean value for
seatbelt installations was 5.1, while the value was 6.7 for LATCH installations. Figure
37 shows the average subject rating of vehicles for LATCH and seatbelt installations.
Installations performed in vehicles A and C had the largest difference in subject
assessment of seatbelt and LATCH installs.
41
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
9
LATCH
Average subject rating
8
Seatbelt
7
6
5
4
3
2
1
0
A
B
C
D
E
F
Vehicle
Figure 37. Average subject ratings of vehicles for LATCH and seatbelt installations.
Subject identification of CRS installation methods
Part of the subject questionnaire involved assessing whether subjects could identify
available methods of CRS installation for a particular vehicle. After each installation, the
subject was asked to mark a vehicle diagram as follows:
Put an S in all the positions where you could install a child seat using the seatbelt.
Put an L in all the positions where you could install a child seat using LATCH.
Put a T in all the positions where you can attach a top tether.
Figure 38 shows a diagram of the correct answers for each vehicle used in the study.
While they vary for each vehicle, all vehicles can use a seatbelt to install a CRS in the
right-front position, and can install with seatbelt, LATCH, and tether in the second-row
outboard positions.
42
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 38. Diagram of options for installing CRS in each test vehicle. S: seatbelt,
L:LATCH, T: tether, *:Center LATCH installation with inboard LA from outboard
positions
Figure 39 shows a distribution of the subject answers for these three seating positions.
For the right-front position, about 20% of subjects correctly answered that a CRS could
43
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
be installed with a seatbelt in that position. Almost 75% incorrectly answered that a CRS
could not be installed in that position. Based on these answers, the message that rear
seating is best for CRS installation is being interpreted that front-seat installation is not
possible. Less than 10% of subjects indicated that it was possible to install CRS in the
front seat using LATCH.
Responses for the two outboard positions were similar but not identical. About 60% of
subjects correctly indicated that seatbelt, LATCH, and tether could be used in the
outboard second-row seating positions. About 5% incorrectly indicated that it was not
possible to use a seatbelt for the installation. About 15% indicated that seatbelt or
LATCH could be used, but did not separately mark that a tether could be used. Almost
10% indicated that only a seatbelt could be used for installation.
100%
Percentage of Subject Responses
90%
80%
Correct
Answer
70%
ST
60%
T
Correct
Answer
50%
Correct
Answer
LT
L
40%
SL
30%
SLT
S
20%
None
10%
0%
1R
2L
2R
Seating Position in Vehicle
Figure 39. Distribution of subject responses regarding available methods for CRS
installation in different vehicle seating positions.
Figure 40 shows the percentage of correct responses regarding allowable methods of CRS
installation in each seating position for each trial. For positions 2L, 2R, and 3R, subjects’
correct response rate improved over the four trials. Results do not show a trend for the
2C, 3L, and 3C seating positions, while they are consistently low for the right-front
seating position.
44
Percentage of subjects correctly
identifying methods of CRS installation
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
100%
90%
80%
Trial
70%
60%
1
50%
2
40%
3
30%
4
20%
10%
0%
1R
2L
2C
2R
3L
3C
3R
Vehicle seating position
Figure 40. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by trial.
Percentage of subjects correctly
identifying methods of CRS installation
The percentage of correct subject answers is shown in Figure 41 for each seating position
according to the number of diagrams in the CRS section of the vehicle manual. For
positions 2L and 2R, the rate of correct answers is inversely proportional to the number
of diagrams. For position 1R, the rate of correct answer is lowest for the two manuals
with the fewest diagrams and similar for the manuals with a higher number of diagrams.
100%
90%
80%
70%
60%
2
50%
5
40%
17
30%
19
20%
47
10%
0%
1R
2L
2C
2R
3L
3C
3R
Vehicle seating position
Figure 41. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by number of diagrams in the CRS portion of the
vehicle manual.
Figure 42 shows the percentage of subjects correctly identifying allowable methods of
CRS installation for each seating position by vehicle. Vehicles D, E, and F have lower
rates of correct identification for position 1R than vehicles A, B, and C. In positions 2L
45
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Percentage of subjects correctly
identifying methods of CRS installation
and 2R, vehicle C has the lowest rate of correct identification compared to the other
vehicles.
100%
90%
80%
70%
A
60%
B
50%
C
40%
D
30%
20%
E
10%
F
0%
1R
2L
2C
2R
3L
3C
3R
Vehicle seating position
Figure 42. Percentage of subjects correctly identifying allowable methods of CRS
installation for each seating position by vehicle.
Evaluation after last install
As a follow-on to the evaluation of CRS instructions and labels performed in Task 1,
after the last installation was performed and assessed, the subject was given the placard
shown in Figure 43 and their response was recorded.
46
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Most people make mistakes when installing car seats.
The biggest problem: getting the car seat in tight enough.



Grab the car seat near the belt path and try to move it.
It should not move more than one inch.
To make it tighter, kneel on the car seat while you tighten
the belt.
The next biggest problem: making the harness snug around the child.


You should not be able to pinch any of the strap.
Use the harness adjuster to tighten it.
Figure 43. Post-test instruction placard.
47
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 44 shows the distribution of actions regarding tightening the installation, almost
all of which were LATCH installs because of the design of the study. In two-thirds of
cases, the subject checked again that it was tight even though it was sufficiently tight. In
13% of cases, the subject did not check the tightness because it was already good. In 9%
of cases, the installation was loose but the subject tried again successfully. For 6% of
cases, the installation was loose and the subject did not try to fix it. For the last 6% of
cases, the installation was loose and the subject was unsuccessful at making it tighter.
Loose, subject
tries again
unsuccessfully
6%
Loose, subject
doesn't do more
6%
Tight, subject
doesn't do more
13%
Loose, subject tries
again succesfully
9%
Tight, subject
checks again
66%
Figure 44. Subject responses to instruction placard on tightening.
Subject responses to the placard regarding snugness of the harness are shown in Figure
45. In 19% of cases, the harness was snug and the subject did not check it again. In 72%
of cases, the harness was snug but the subject checked again. In 9% of cases, the harness
was loose but the subject was unable to adjust it to be snug.
48
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Loose, subject tries
again
unsuccessfully
9%
Snug, subject
doesn't do more
19%
Snug, subject
checks again
72%
Figure 45. Subject responses to instruction placard on harness snugness.
49
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Discussion
Summary and comment on results
Installation tightness
The SAE Child Restraints Committee has developed draft recommended practices to
reduce LATCH incompatibilities between CRS and vehicle seating positions. One of the
factors considered important by this group is the location of lower anchorages above
seating surface, where the recommended distance between the installed CRF fixture and
the seating surface is 50 mm or less. Another SAE recommendation is that the force
required to attach LATCH belt connectors to lower anchorages be less than 15 lb. The
vehicles selected for this study were chosen to differ on these factors. Two vehicles have
lower anchorages located above the seating surface (B and D). Two vehicles had
attachment forces below 15 lb, two were near 15 lb, and two were above 15 lb. In
addition to the SAE recommended practices, there is a perception that visible lower
anchorages may improve usability compared to lower anchorages that are located within
the vehicle seat.
When assessing the tightness of the CRS installations (the most frequent type of misuse),
installation performance did not correspond with any of these three recommendations.
The force required to attach the lower anchorages was not a significant predictor of
installation tightness. Vehicles with visible lower anchorages did not have improved
installation tightness compared to those with lower anchorages buried within the vehicle
seat. However, as discussed below, these two factors did positively affect correct use of
lower anchorages.
In addition, the two vehicles with lower anchorages located above the seating surface had
higher rates of tight installation for both LATCH and seatbelt installations compared to
vehicles with lower anchorages at the seating surface. The two vehicles with the lower
anchorage above the seating surface have a feature called a bightline waterfall (shown in
Figure 9). Because the two vehicles with this feature had improved rates of tight
installation for both LATCH and seatbelt installations, but had different levels of seat
stiffness, seat contour, and buckle location, it suggests that the shape of the vehicle seat
with this feature may have a greater effect on installation tightness than the location of
the lower anchorages. The improved installations with these vehicle seats is at odds with
the SAE CRS committee recommendation to have the lower anchorages close to the
seating surface to allow good contact between the CRS and vehicle seat.
In this study, seatbelt buckles that were located closer to the seat bight led to a
substantially higher rate of tight seatbelt installations compared to buckles located
forward of the bight (83% vs. 35%). However, the location of seatbelt buckles did not
have an effect on LATCH installations, indicating that subjects are able to access lower
anchorages even if the seatbelt buckle is located nearby. In addition, subject
questionnaires responses indicated that subjects did not find the seatbelt buckles to cause
a problem when performing LATCH installs.
50
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Only two CRS were used in Task 2 testing. CRS C10, which uses the same belt path for
routing the seatbelt and the LATCH belt, had the same 60% rate of tight installation for
seatbelt and LATCH installs. CRS C6 has the vehicle belt path located slightly higher
than where the LATCH belt is attached. With C6, the rate of tight installations with
LATCH was twice the rate of tight installations with seatbelt (80% vs. 40%). The
differences in performance with these two CRS over a range of vehicles indicate that belt
path location on the CRS plays a key role in achieving tight installations in vehicles. On
the FMVSS 213 buck, the geometry of the lap belt anchors and the LATCH lower
anchorages are similar. In vehicles, lap belt anchorages are almost always located more
forward than the LATCH anchorages. If the typical lap belt geometry implemented in
vehicles is different from that found on the FMVSS 213 test buck, CRS manufacturers
may be optimizing the design of their belt paths for unrealistic conditions.
The vehicles tested were selected to provide a variety of head restraint conditions,
because problems with head restraints interfering with CRS installation have been
reported by CPS technicians in the field. Since the majority of the vehicles used in this
study were manufactured before the phase-in of the new FMVSS 202 head restraint
requirements, it is not known whether this finding applies to compliant designs.
However, no factors related to head restraints were significant predictors of installation
tightness. Subject questionnaire responses did indicate that the head restraints in the
different vehicles had different effects on the subjects’ perception of the installations,
even if they did not affect installation tightness.
Lower anchorage use
Subjects were assessed as correctly using the lower anchorages if they attached the
LATCH connectors to the appropriate lower anchorages and attached the connectors
correctly (i.e. not upside-down). Overall, subjects attached the LATCH belt correctly in
90% of LATCH installations. Subjects were less successful in correctly attaching the
LATCH belt if the lower anchorages were buried rather than visible, or if the force
required to attach the connectors was high. These findings support two of the SAE
recommended practices for LATCH usability. However, correct use of the lower
connectors was relatively high for vehicles not meeting the SAE recommendations: 88%
for buried anchors and 79% for high lower anchorage force.
The main selection criteria for the two CRS used in Task 2 were to provide examples of
hook-on and push-on LATCH connectors. Consistent with the results of Task 1, subjects
were more likely to correctly attach to the lower anchorages using the push-on style of
LATCH connector rather than the hook-on style.
Tether use
Tether use was evaluated in two parts: correct attachment and tightening. The tether was
used correctly in only 30% of all installations, and not used in 26% of all installations.
Subjects used the tether in 85% of LATCH installations and only 48% of seatbelt
51
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
installations. When it was used, it was used correctly in 45% of LATCH installations and
30% of seatbelt installations. The most common error in tether use was routing the tether
strap over the head restraint, rather than under it or removing the head restraint to
accommodate the CRS.
The location and conspicuity of the tether anchorage had a major effect on tether use.
The tether was more likely to be used in both LATCH and seatbelt installations in sedans
where the tether anchorage is located on the package shelf and visible from the point of
installation. Tether use was lower when the anchorage was located on the vehicle
seatback, where it is often found in minivans and crossover vehicles. A factor that may
be contributing to this difference is that the tether anchorages located in three of the four
sedans were located under clearly marked doors as shown in Figure 46, whereas all three
tether anchorages located on the seatbacks were not marked as shown in Figure 47.
Subjects overwhelmingly indicated that tether anchorages were more difficult to locate in
the three vehicles where they were mounted on the seatback. As a result, subjects were
twice as likely to use the vehicle manual in these vehicles compared to sedans.
Figure 46. Representative tether anchorage under marked door in sedans.
Figure 47. Representative unmarked tether anchorage often found in
minivans/crossover vehicles.
52
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
One of the measures used to describe head restraint geometry was the fore-aft distance
between an estimated seat H-point and the base of the head restraint in its lowest
adjustment position. The vehicles with the base of the head restraint more forward had
higher rates of correct tether attachment than the vehicles with the head restraint more
rearward, while the opposite is true when evaluating rates of tight tether installation.
There is a perception that more forward head restraints are bad for CRS installation, but it
is not as straightforward as that. More research is needed to determine if more forward
head restraints permit greater visibility and thus use of the tether anchorages.
The tether anchorage zone is defined in FMVSS 225 relative to an R-point determined by
using the H-point machine. A problem reported from the field is that some tether
anchorages in the most forward portions of this zone do not permit enough space to allow
adequate tightening of the tether. An example of this condition is shown in Figure 48.
The tether wrap around distance was measured in each vehicle relative to an estimated Rpoint. In the vehicle with a tether wrap around distance of 210 mm, all of the tether
installations were tight. In the vehicle with the tether wrap around distance of 180 mm,
less than 40% of the installations were tight, and could only be achieved by incorrectly
routing the tether over the head restraint as shown in Figure 49. Thus the threshold for
sufficient space for tightening the tether (based on these two CRS) likely lies between
180 and 210 mm. When the other 14 convertible child restraints were installed in the
vehicle with the wrap around distance of 210 mm, 5 others could not be tightened.
Figure 48. Tether anchorage located too close to allow space to tighten tether.
53
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Figure 49. Incorrectly routed but tight tether installation.
The minimum wrap around distance required to tighten tether depends on the length of
tether hardware (hook plus adjustment mechanism) as well as the attachment location of
the tether on the child restraint. Two child restraints with the same length tether
hardware could require different minimum wrap around distances in the vehicle if they
were attached at different heights on the child restraint. Additional work is needed to
develop a specification in FMVSS 225 for minimum vehicle wrap around distance that is
compatible with a specification in FMVSS 213 for a maximum tether hardware length.
Seatbelt installations
As mentioned previously, subjects had tight installations using the seatbelt twice as often
when the seatbelt buckle was located at the bight rather than forward of the bight.
Subjects were also more likely to lock the seatbelt when it was at the bight rather than
forward of the bight. It is not clear if the subjects were unable to tighten the belt with a
forward buckle, and thus did not bother to lock the seatbelt, or if the locking options were
less evident in vehicles with buckles forward of the bight. When locking the belt, most
subjects used the method they were directed to use by the CRS manual. One of the
vehicles had a locking latchplate (where no action is needed to lock the belt), whereas the
rest had switchable retractors. In the four seatbelt installations performed in the vehicle
with the locking latchplate, the seatbelt was locked in all cases and the CRS was tight in
three of four cases.
Instruction manual use
Subjects used the vehicle manual in 38% of installations and consulted the CRS manual
in 88% of installations. The more difficult-to-find tether anchorages in the
minivans/crossover vehicles doubled the frequency of subjects consulting the vehicle
manual.
Subjects gave different ratings for each vehicle as to whether the content of vehicle
manual agreed with the CRS manual. (Only ratings from subjects who had used the
vehicle manual were included.) One of the vehicle manufacturers had consistent ratings
54
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
from subjects on two of its vehicles, while the other manufacturer with two vehicles had
the highest and lowest ratings.
Subjects also gave a variety of scores to each vehicle with regard to understanding the
CRS installation instructions included in each vehicle manual. The two shortest manuals
with the fewest diagrams had the highest scores, while the manual with the longest CRS
section (and the most diagrams) received the lowest rating on ease of understanding. The
manual that would likely be considered the most thorough was the one rated most
difficult to understand.
Installation times increased substantially if subjects used the vehicle or CRS manual
compared to those who did not. Experienced subjects were less likely to use the vehicle
manual, especially those with lower education levels. Subjects who did not use any
manual had more installation errors than subjects who used the vehicle manual, CRS
manual, or both.
Education/experience
Education and experience affected few CRS installation factors. Rate of tight
installations was somewhat lower for lower education subjects, while rate of tight
tethering (once it was used) was higher for lower education subjects. In Task 1, the effect
of education on tight installations had the same trend, but was not a factor in achieving a
tight tether. Inexperienced subjects were more likely to use the vehicle and CRS
manuals, but this was not found in Task 1.
Subject assessments
Data from the post-installation questionnaire identified only a few areas where the
average response differed among vehicles. Since their installation performance varied
among vehicles, this indicates that subjects are not aware of when they are making errors
during CRS installation. Overall, subjects rated themselves as correctly installing the
child restraints (average score of 4.1/5 corresponding to agreement with “I attached the
child seat to the vehicle correctly”). However, when reviewing subject performance,
over half made two or more errors on key installation points (CRS tight, harness snug,
buckle correct, tether correctly attached, tether tight, LATCH belt correct/seatbelt
locked). These results show a disconnect between subject self-evaluation and actual
performance.
Subjects did not identify the vehicle seat stiffness or vehicle seat shape as contributing to
the difficulty of the installation. They also did not find that the seatbelt buckles got in the
way of using LATCH. The only vehicle feature noted as making it harder to install was
the head restraint. Table 7 shows pictures of each head restraint in order from hardest to
easiest as rated by subjects, together with the instructions for routing the tether relative to
the head restraint. The head restraint considered most challenging to CRS installation
was the one with a fixed design. Vehicles D and A have similar head restraint designs,
but the tether anchorages for vehicle D are located on the package shelf while those for A
55
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
are on the seatback, which may contribute to the different ratings despite the similar head
restraint design. The same may be true for vehicles B and F.
Table 7.
Vehicle head restraints with average subject rating of difficult (hardest to
easiest)
Vehicle E (3.4)
Route belt to outboard side
Vehicle D (3.0)
Route under
Vehicle B (2.9)
Route under, or around if
tether cannot be tightened
Vehicle A (2.4)
Route under
Vehicle C (2.4)
Remove HR
Vehicle F (2.1)
Route under
Subject ratings on most questions addressing LATCH installation, seatbelt installation,
and CRS instructions were similar for all vehicles and generally positive. As discussed
previously, the two questions having the widest range of responses were the ease of
finding the tether anchorage and understanding the vehicle manual. A preference for
LATCH installation was expressed by 80% of subjects at the end of their session.
Subject understanding of CRS installation options
Subjects were asked to indicate on a diagram the allowable methods of CRS installation
at each seating position in the vehicle. About three-quarters of subjects incorrectly stated
that child restraints could not be installed in the front seat. This error likely results from
interpreting vehicle manual instructions prohibiting installation of rear-facing CRS in the
front seats as covering all CRS installations, or that subjects have received the child
passenger safety message that kids should sit in the back seat.
Based on both subject responses and their performance on installations, many people do
not realize that tethers can be used when the seatbelt is the primary method of installing
the child restraint. This suggests an opportunity for improved education.
56
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Subject responses on identifying CRS installation options for each seating position
tended to be correct most often for the 2nd row right seating position, where they were
asked to install the CRS in most vehicles. They also had a high rate of correct answers in
the 2nd row left position, which for all of the vehicles tested, had the same CRS
installation options as the 2nd row right position.
Post-test evaluation of subject feedback
Results of Task 1 and Task 2 both indicated that subjects did not recognize when they
made installation errors, because their assessments of how they did had minimal
correlation with how they performed. As a quick experiment to determine if providing
post-installation feedback would lead subjects to improve their installation, subjects were
given the placard shown in Figure 43 after performing their last installation and their
response was tabulated. The placard focuses on correcting the two most frequent
misuses: loose installation, and loose harness. After giving the subjects the placard, 80%
went back and checked their installations. The subjects improved the tightness of the
installation in three of the seven cases where it was not tight, but none of the subjects
who needed to make the harness tighter were able to figure out how to do so.
Comparison to Task 1 results
Compared with subjects in Task 1 where four subjects quit the study and seven did not
finish all four trials, no subjects withdrew from this phase of the study, and only one did
not finish all four trials. Factors thought to contribute to these differences are:
1) Dummy installation was much less complicated, because the harness height was
already adjusted to fit the dummy. This seemed to reduce subjects’ frustration
levels, making them less likely to give up once they reached the CRS installation
portion of the trial.
2) Installations were all forward-facing. In Task 1, although performance did not
generally vary between forward-facing and rear-facing installations, rear-facing
installations took twice as much time. Achieving the correct installation angle
rear-facing was shown to be more challenging than in forward-facing
installations.
3) The CRS selected for this study had the highest rates of correct installation in
Task 1.
Table 8 compares misuse rates on key variables between Task 1A and Task 2. In Task 2,
misuse rates were substantially lower for all items except correct buckling and the CRS
being installed with both seatbelt and LATCH. The decrease in incorrect belt routing and
incorrect recline angle likely results from performing only forward-facing installations.
The decrease in loose harness rate likely results from setting the harness to the correct
height for the dummy prior to testing, as well as choosing two of the easiest-to-use CRS
from Task 1 for use in Task 2.
57
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
Table 8. Comparison of misuse rates (percentages) in Task 1A and Task 2 testing
Observed Misuse
Task 1a
Task 2
(n=116)
(n=132)
Study design
Incorrect belt routing
Incorrect recline angle
Loose CRS install
CRS installed with both
seatbelt and LATCH
Harness incorrectly buckled
Loose harness
16 CRS
2 CRS
1 vehicle
7 vehicles
50/50 LATCH/Seatbelt 75/25 LATCH/Seatbelt
50/50 RF/FF
All FF
17
2
22
2
72
33
4
7
9
48
8
26
Rates of loose CRS installation in Task 1 were more than double the rate of loose CRS
installation from Task 2. This partly results from choosing easy-to-use CRS, and partly
from the vehicle factors. Vehicle G was chosen for Task 1 because it had vehicle factors
expected to result in uncomplicated installations such as flat seat contours, visible lower
anchorages, and seat belt anchorages near the bight. However, these features did not turn
out to be as uncomplicated as anticipated. Comparing the installation errors using CRS
C6 and C10 from Task 1 testing to that of the other vehicles in Task 2, Vehicle G had the
lowest rate of tight LATCH installations and the second lowest rate of tight seatbelt
installations. However, some of this difference may result from fewer trials with these
two CRS in Task 1 testing compared to the number of trials in each vehicle in Task 2
testing.
The experimenter noticed a marked reduction in frustration levels of subjects during this
phase of testing compared to the previous testing. She believes that it contributed
substantially to the improved performance of subjects. This may suggest that improving
ease-of-use in one area of child restraint design (such as making harnesses easier to
adjust) may indirectly reduce installation errors in other aspects of installation (such as
tightly installing the child restraint.)
Limitations
Because the project constraints only allowed testing with six vehicles, the number of
vehicle features that could be examined was limited. In addition, only 30 vehicles were
evaluated as potential candidates for testing because of time and budget constraints.
Although efforts were made in analysis to evaluate each feature separately, actual CRS
installation is affected by all features together.
Only two convertible CRS were used during testing. They were selected to be relatively
easy to use based on results of Task 1 testing, and had different external contours, belt
58
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
paths, and LATCH belt connectors. However, findings based on these two CRS may not
apply to all CRS. In addition, all installations were performed forward-facing, because
this meant all installations could be used to examine tether use, and more degrees of
freedom in the experimental design could be applied to vehicle factors. Findings may not
apply to rear-facing installations.
Some of the draft SAE procedures on LATCH compatibility were used to assess the
potential vehicles for testing. As written, the method for measuring the force required to
attach the LATCH belt connectors to the lower anchorages is not very repeatable,
because the attachment force is highly dependent on the angle at which the gauge is
applied, and it is physically impossible to use the same measurement angle in all vehicles.
We have confidence in the force categories used as predictors in this report because they
are based on multiple measures in the vehicles. However, further refinement of the
procedure is needed before it could be used in assessing vehicles.
Because of time constraints resulting from renting the test vehicles for a limited period of
time, we were not able to take the time to recruit as wide of an age range of subjects in
each education/experience group. Thus the X0L group (no experience, have not gone to
college) is mainly composed of recent high school graduates, almost all of whom will be
attending college in the fall. This makes them younger than the X0L groups in the prior
phases of testing, which included more older subjects who never attended college.
Another limitation of the subject pool is that all subjects could speak and read English,
with their performance likely better than subjects who would not have the language skills
to understand the labels and manuals.
There was a significant effect on tether use between vehicles with tether anchorages
located on the package shelf, most of which were clearly marked, and those located on
the vehicle seatback, which were all unmarked. The design of the study is not able to
determine whether it was the location or the marking (or some combination) that led to
the improved tether use in sedans.
An effort was made to test vehicles with fabric rather than leather seats, because field
reports generally indicate that leather seats contribute to difficult CRS installation.
Unfortunately, vehicle model A was only available with leather seats, and this factor may
have contributed to its performance.
The results of both Task 1 and Task 2 of this study have indicated that opinions of
experts in child passenger safety on what makes CRS installation hard or easy does not
necessarily lead to poor or good performance by regular CRS users. Some particular
examples:
 The vehicle chosen for testing in Task 1, because it was considered to have easy-touse vehicle features, had among the lowest rates of tight installation compared to the
other vehicles in Task 2.
 Two of the key recommendations by the SAE CRS committee (avoid lower
anchorages above the seating surface and limit force required to attach lower
anchorages) did not have a significant effect on installation tightness, although they
slightly improved rates of correctly attaching the LATCH belt.
59
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report

Child restraint manuals and labels designed according to human factors principles
recommended by experts (evaluated in Task 1) did not have a significant effect on
improving CRS installation.
These findings are cause for some concern, as the NHTSA ease-of-use rating system and
ISO rating systems of CRS ease-of-use have primarily been based on consensus by child
passenger safety experts. Testing with non-expert volunteers should be performed to
validate any recommendations by experts intended to improve ease of child restraint
installation. Narrowing the range of recommendations to those factors demonstrated to
affect CRS installations may improve the responsiveness to those recommendations by
CRS and vehicle manufacturers.
Recommendations
The results of volunteer testing to evaluate effects of vehicle features on CRS installation
errors support the following recommendations:







Clear marking of tether anchorage locations should be required. Because existing
ISO recommendations for tether anchorage markings have been adopted by some
vehicle manufacturers, use of the ISO specifications is suggested. The substantial
safety advantages of tether use, combined with the current finding that hard-to-find
anchorages dramatically reduce tether use, suggest that a strong emphasis on
improving tether anchorage marking would yield safety benefits.
Further testing should be conducted to identify an acceptable minimum tether wrap
around distance for the tether anchorage zone that is compatible with tether hardware
lengths and attachment locations.
Because forward seatbelt buckle locations had lower rates of tight installation than
buckles anchored more rearward, the effect of lap belt anchorage location on CRS
installation should be assessed. Implications for both FMVSS 210 and FMVSS 213
should be considered.
Lower anchorages requiring less force to attach lower connectors, as well as visible
lower anchorages, led to improved rates of correctly attaching lower anchorages.
The potential for negative interaction between head restraints and child restraints
should be monitored as more vehicles are redesigned to meet FMVSS 202a
requirements.
Education efforts targeting increasing LATCH use should also emphasize the benefits
of using a tether with a seatbelt installation.
New requirements intended to improve CRS ease-of-use should be validated with
testing by non-experts.
60
Effects of Vehicle Features on Child Restraint Installation Errors: Task 2 Final Report
References
Bilston, LE, Yuen, M, Brown J (2007). Reconstruction of crashes involving injured child
occupant: the risk of serious injuries associated with sub-optimal restraint use may be
reduced by better controlling occupant kinematics. Traffic Injury Prevention 8:47-61.
Carlsson, G, Norin, H, Ysander, L, (1991). Rearward facing child seats – the safest car restraint
for children? Accident Analysis and Prevention 23:2-3:175-182.
Code of Federal Regulations (2005) Title 49, Transportation, Part 571.213; Child restraint
systems. National Archives and Records Service, Office of the Federal Register
Washington, D.C.
CDC (2007). http://www.cdc.gov/injury/wisqars/pdf/Death_by_Age_2007-a.pdf
Decina, LE, Lococo, KH, (2007). Observed LATCH use and misuse characteristics of child
restraint systems in seven states. Journal of Safety Research 38:271-281.
Eby, DW, Kostyniuk, LP (1999). A statewide analysis of child safety seat use and misuse in
Michigan. Accident Analysis and Prevention 31:555-566.
International Standards Organization (2008) Road Vehicles – Methods and Criteria for Usability
Evaluation of Child Restraint Systems and Their Interface With Vehicle Anchorage
Systems – Part 1 Vehicles and Child Restraints Equipped with ISOFIX Anchorages and
Attachments. CD 29061-1, Geneva, Switzerland.
Johnston, C, Rivara, FP, Soderberg, R (1994) Children in car crashes: analysis of data for injury
and use of restraints. Pediatrics 93:960-965.
Klinich, K. D., Manary, M. A., Flannagan, C. A. C., Ebert, S. M., Malik, L.A., Green, P. A., and
Reed, M. P. (2010). Labels, instructions, and features of convertible Child Restraint
Systems (CRS): Evaluating their effects on CRS installation errors. UM-2010-xx.
Koppel, S, Charlton, JL (2009). Child restraint system misuse and/or inappropriate use in
Australia. Traffic Injury Prevention 10:302-307.
Lane, WG, Liu CL, and Newlin, E (2000) The association between hands-on instruction and
proper child safety seat installation. Pediatrics 106(4):924-928.
National Highway Traffic Safety Administration (2006). Consumer Information Rating Program
for Child Restraints Docket NHTSA-2006-25344, US Department of Transportation,
Washington DC.
National Highway Traffic Safety Administration (2009). Traffic Safety Facts 2008, DOT HS 811170, National Center for Statistics and Analysis, Washington DC.
Ruta D, Beattie T, Narayan, V (1993). A prospective study of non-fatal childhood road traffic
accidents: what can seat restraint achieve? Journal of Public Health Medicine 15:88-92.
SAE Child Restraints Systems Committee (2009). Guidelines for Implementation of the Child
Restraint Anchorage System or LATCH System in Motor Vehicles and Child Restraint
Systems. Draft Recommended Practice.
Subramanian R (2003). Motor Vehicle Traffic Crashes as a Leading Cause of Death in the United
States, 2001 DOT HS 809 695 National Center for Statistics and Analysis, Washington
DC.
Stewart, D. D., Anderson, M., Lang, N. J., Rose, K. and White, L. (2009). The LATCH Manual:
Using lower anchors and tethers for child restraints. Sixth Edition. Safe Ride News
Publications.
61
Appendix A: Summary of FMVSS 225
Appendix A
Overview of Federal Motor Vehicle Safety Standard (FMVSS) 225
62
Appendix A: Summary of FMVSS 225
FMVSS 225 requires that instructions for use of LATCH be provided by the vehicle
manufacturer in the vehicle manual. For all vehicles manufactured after September 1,
2002, tether anchorages are required in three rear seating positions in any vehicle with
three or more rear seating positions, including at least one location that is not outboard.
Lower anchorages are required in at least two positions in vehicles with at least three rear
seating positions. If a vehicle has three or more rows of seats, at least one tether
anchorage and one set of lower anchorages must be in the second row. If a vehicle has
two or fewer rear seating positions, they must all have tether anchorages and lower
anchorages. If there is no rear seating position, each front-passenger seating position
must have a tether anchorage. Tether and lower anchorages must be accessible at all
times that the associated vehicle seating position is available for use (i.e. when the seat is
not stowed for cargo).
Tether anchorages must accept a tether hook that complies with FMVSS 213, be
accessible to the consumer with nothing more than a screwdriver, and be sealed to
prevent exhaust fumes from entering the passenger compartment. Tether anchorages
must be positioned in a zone defined relative to the seating reference point (SgRP) for the
associated seating position. If it is not possible to install tether anchorages in this zone,
tether-routing devices can be installed that redirect the tether to an anchorage outside the
zone.
FMVSS 225 uses two types of Static Force Application Devices (SFAD) and a Child
Restraint Fixture (CRF) in test procedures. To assess tether anchorage strength, SFAD 1
is used to apply static load to the tether anchorage in seating positions without lower
anchorages, while SFAD 2 is used in seating positions with lower anchorages. The
SFAD is attached to the vehicle seat and the tether strap is secured to the anchorage.
Force is applied through point X on the SFAD at an angle of 10 +/-5 degrees above
horizontal. The procedure requires application of a force of 15,000 N within a 24 to 30
second loading event, and a sustained peak loading of at least 1 second. The anchorage
must not separate from the vehicle structure to which it is attached, but there are no
maximum hardware displacements specified. Tether anchorages in the same row must be
tested simultaneously if they are located at seating positions whose centerlines are 400
mm or greater apart.
For evaluating strength of lower anchorages, SFAD 2 is attached to the lower anchorages
(without using a tether) and a force of 11,000 N is applied in the fore-aft direction +/- 10
degrees of horizontal within a loading period of 24 to 30 seconds and maintained at peak
load for at least 1 second. For lateral loading, the procedure is repeated except that the
maximum load is 5000 N and the pull angle tolerance is +/- 5 degrees. The strength
requirement is that point X on the SFAD 2 cannot displace more than 175 mm under
fore-aft loading or more than 150 mm under lateral loading. Lower anchorages whose
seating position centerlines are 400 mm or more apart must meet the requirements when
tested simultaneously.
Other requirements for lower anchorages include specifications for the diameter, shape,
length, and depth of the anchor bars that comprise the lower anchorages. The bars must
be rigidly attached to the vehicle such that they will not deform more than 5 mm when
100 N is applied in any direction. In addition, there are requirements for the pitch, roll,
63
Appendix A: Summary of FMVSS 225
and yaw of a Child Restraint Fixture (CRF) when attached to the lower anchorages, as
well as a specification for the fore-aft location of the lower anchorages relative to the
CRF and vehicle SGRP. The CRF must be able to fit in the vehicle seat and be attached
to the lower anchorages.
There are specifications for the conspicuity of the lower anchorage locations. Either the
lower anchorage bars, a guide device to reach the bars, or a label indicating the lower
anchorage locations must be visible from a specified angle. The label must include a 13mm-diameter circle, have some indication of lower anchorages that is defined in the
vehicle manual, and be permanently attached to the vehicle. The regulation also contains
requirements for written LATCH instructions in the vehicle manual. Instructions must be
in English, state which seating positions have lower anchorages and tether anchorages,
and describe the markings used to designate the anchorage locations. The instructions
must also include a procedure for attaching the tether strap.
64
Appendix B: Photos of test vehicles
Appendix B:
Photos of test vehicles
65
Appendix B: Photos of test vehicles
A: 2010 Ford Flex
B: 2008 Honda Civic
C: 2008 Toyota Sienna
D: 2010 Ford Fusion
E: 2008 Dodge Avenger
F: 2008 Chrysler Pacifica
Photos of rear seats of proposed test vehicles.
66
Appendix B: Photos of test vehicles
A: 2010 Ford Flex
B: 2008 Honda Civic
C: 2008 Toyota Sienna
D: 2010 Ford Fusion
E: 2008 Dodge Avenger
F: 2008 Chrysler Pacifica
Photos showing LATCH markings/locations of each vehicle.
67
Appendix B: Photos of test vehicles
A: 2010 Ford Flex
B: 2008 Honda Civic
C: 2008 Toyota Sienna
D: 2010 Ford Fusion
E: 2008 Dodge Avenger
F: 2008 Chrysler Pacifica
68
Appendix B: Photos of test vehicles
A: 2010 Ford Flex
B: 2008 Honda Civic
C: 2008 Toyota Sienna
D: 2010 Ford Fusion
E: 2008 Dodge Avenger
F: 2008 Chrysler Pacifica
69
Appendix C: Installation Evaluation Forms and Test Protocol
Appendix C
Installation Evaluation Forms and Test Protocol
70
CRS Installation Evaluation Form (completed by experimenter)
Subject ID:
Vehicle: A B C D
Installation number: 1 2 3 4
CRS: C6 C10
Configuration: RF FF Method: L SB Both
Installed position: 1R 2L 2C 2R 3L 3C 3R
Start time:
End time:
Date:
Evaluator:
Yes No NA Comment
Did subject install CRS in directed position?
Did subject install with both LATCH and seatbelt?
Did subject use vehicle manual?
Did subject use child restraint manual?
Did subject adjust front seat (where)?
Does CRS pass 1” movement test (measure)?
Is CRS touching front seatback?
Did subject use noodles or towels (what, #)?
Is recline angle set correctly?
Is belt routed through correct path?
Is belt twisted?
Did subject use locking clip?
Did subject use lock-offs correctly?
Did subject use retractor to lock belt?
Did subject use locking latchplate to lock belt?
LATCH attached to correct lower anchors?
Lower connectors attached appropriately?
Is tether hook attached to anchorage?
Is tether hook attached correctly?
Is tether routed correctly wrt headrests?
Is tether tight? (measure)
Is tether stored?
Is LATCH stored?
Did subject remove vehicle headrest?
Did subject buckle harness correctly?
Is harness snug (measure)?
69
Testing Script / Protocol
Definitions of assessed items
Did subject install
CRS in directed
position?
Did subject install
with both LATCH
and seatbelt?
Did subject use
vehicle manual?
Subject locates and opens vehicle manual
Did subject use
child restraint
manual?
Subject locates and opens child restraint manual
Did subject adjust
front seat (where)?
Does CRS pass 1”
movement test
(measure)?
Experimenter grasps child restraint near belt path and determines if
it moves less than 1” in any direction. For quantitative
measurement, masking tape is used to mark the location of the CRS
relative to the seat cushion. The experimenter applies a horizontal
force of 40 lb to the CRS near the belt path and measures the
amount of displacement.
Is CRS touching
front seatback?
Did subject use
noodles or towels
(what, #)?
Is recline angle set
correctly?
Is belt routed
through correct
path?
For forward-facing installations used in this study, the subject uses
the forward-facing recline position.
The belt is routed through the forward-facing path.
Is belt twisted?
The belt is not flat and/or has twists in it.
Did subject use
locking clip?
Identify whether subject used the supplemental locking clip
provided with the CRS to lock the seatbelt.
If CRS comes with lockoffs for locking the seatbelt, the subject
used them as indicated in the child restraint manual.
Did subject use
lock-offs correctly?
Did subject use
retractor to lock
belt?
If subject locked seatbelt by using the switchable retractor (when
available) in locked mode.
Did subject use
locking latchplate to
lock belt?
If subject used locking latchplate (when available) to lock seatbelt.
LATCH attached to
correct lower
anchors?
The lower connectors are attached to the correct vehicle hardware.
Lower connectors
The lower connectors are fully engaged in the right orientation with
70
Testing Script / Protocol
attached
the LATCH webbing flat.
appropriately?
Is tether hook
attached to
anchorage?
Tether hook attached to the correct vehicle hardware.
Is tether hook
attached correctly?
The tether hook is fully engaged and in the correct orientation.
Is tether routed
correctly wrt
headrests?
The tether is routed with respect to the head restraints in the manner
specified by the vehicle owners manual.
Is tether tight?
(measure)
Pinch the slack in the tether webbing into a loop and measure the
height of the loop. Tether is tight if height of loop is 5 mm or less.
If tether is not used, is it stored in the location provided on the child
restraint.
If LATCH belt is not used, is it stored in location provided on the
child restraint?
Is tether stored?
Is LATCH stored?
Did subject remove
vehicle headrest?
Did subject buckle
harness correctly?
Is harness snug
(measure)?
Pinch the slack in the harness webbing into a loop and measure the
height of the loop. Harness is snug if webbing cannot be pinched.
71
Testing Script / Protocol
CRS should be set up with harness position closest to dummy’s shoulders. Rest of
components should be in “out-of-the-box” configuration, including:
Recline
Tether storage
LATCH storage
Instruction storage
Introduction
Thank you for coming in today. We’re doing a study on how people install child seats,
and we are going to ask you install a child seat four times, once in each vehicle.
You can use the instructions for the child seat and the vehicle.
Let me know each time when you are done – I will take some measurements and you will
answer some questions, then we will go on to the next child seat.
We will videotape some of the installations. When we do, we would like you to talk
about what you are doing and thinking.
You might want to remove your jewelry (watch, large rings, etc.).
Please remember that most people make mistakes when installing child seats. We want
you to do your best, but not get frustrated. We are testing the child seats and vehicles,
not you. I am not allowed to help you install the child seat. If you ask me a question, I
might not be able to give you a clear answer.
This is a consent form for you to be in our study. Please look through it and let me know
if you have any questions. I will give you a copy of the form to keep.
We would also like you to fill out this ethnicity form. You can still participate if you do
not want to fill out this form.
Give subject consent form to read and sign; give subject ethnic/race form to fill out.
The top of this cart has tools you can use for installing the child seat. This is your baby
for today. He is 18 months old and weighs 25 pounds and we have set up the harness on
the car seat so he will fit in it. Here are the instructions for the child restraint, and the
vehicle instructions are stored in the glove compartment (or where they are).
Pool noodles and flat screwdriver will be on test cart.
Installation #1
Please install this seat ______________ facing for this child at the (right rear/center)
______________ position in the vehicle. We would like you to try installing the car seat
using (LATCH/the seatbelt).
 By forward facing, I mean the child is facing the same direction as the driver.
 By rear facing, I mean the child is facing the trunk.
Point subject towards first child restraint/vehicle to be tested.
Record start time of installation.
72
Testing Script / Protocol
If subject cannot install using the method in that position, try having them install using
the other method.
Record end time of installation.
Give subject questionnaire – direct them to fill it out behind a screen so they can’t view
the experimenter checking installations.
Assess installation using check form.
If you want to look at the labels or instructions to answer the questions, let me know.
If so, experimenter will pause assessment while subject reviews labels on installed child
seat. Experimenter can answer questions about filling out the form, such as identifying
CRS or vehicle features (e.g. this thing is the harness clip, this is the tether anchorage).
Installation #2
Please install this seat ______________ facing for this child at the (right rear/center)
______________ position in the vehicle. We would like you to try installing the car seat
using (LATCH/the seatbelt).
 By forward facing, I mean the child is facing the same direction as the driver.
 By rear facing, I mean the child is facing the trunk.
Point subject towards second child restraint to be installed.
Record start time of installation.
If subject cannot install using the method in that position, try having them install using
the other method.
Record end time of installation.
Give subject questionnaire – direct them to fill it out behind a screen so they can’t view
the experimenter checking installations.
Assess installation using check form.
If you want to look at the labels or instructions to answer the questions, let me know.
If so, experimenter will pause assessment while subject reviews labels on installed child
seat. Experimenter can answer questions about filling out the form, such as identifying
CRS or vehicle features (e.g. this thing is the harness clip, this is the tether anchorage).
Installation #3
Please install this seat ______________ facing for this child at the (right rear/center)
______________ position in the vehicle. We would like you to try installing the car seat
using (LATCH/the seatbelt).
 By forward facing, I mean the child is facing the same direction as the driver.
 By rear facing, I mean the child is facing the trunk.
Point subject towards third child restraint/vehicle to be tested.
Record start time of installation.
73
Testing Script / Protocol
If subject cannot install using the method in that position, try having them install using
the other method.
Record end time of installation.
Give subject questionnaire – direct them to fill it out behind a screen so they can’t view
the experimenter checking installations.
Assess installation using check form.
If you want to look at the labels or instructions to answer the questions, let me know.
If so, experimenter will pause assessment while subject reviews labels on installed child
seat. Experimenter can answer questions about filling out the form, such as identifying
CRS or vehicle features (e.g. this thing is the harness clip, this is the tether anchorage).
Installation #4
Please install this seat ______________ facing for this child at the (right rear/center)
______________ position in the vehicle. We would like you to try installing the car seat
using (LATCH/the seatbelt).
 By forward facing, I mean the child is facing the same direction as the driver.
 By rear facing, I mean the child is facing the trunk.
Point subject towards fourth child restraint/vehicle to be tested.
Record start time of installation.
If subject cannot install using the method in that position, try having them install using
the other method.
Record end time of installation.
Give subject questionnaire – direct them to fill it out behind a screen so they can’t view
the experimenter checking installations.
Assess installation using check form.
If you want to look at the labels or instructions to answer the questions, let me know.
If so, experimenter will pause assessment while subject reviews labels on installed child
seat. Experimenter can answer questions about filling out the form, such as identifying
CRS or vehicle features (e.g. this thing is the harness clip, this is the tether anchorage).
Thank you for being in our study today.
Please fill out this payment form so we can pay you.
If subject decides to drop out of the study, pay $12/hr rate for the participation so far.
If subject does not complete all four installations within 3 hours, they can stay longer if
possible or just finish the third installation.
74
Testing Script / Protocol
Questions
If subject can’t find instructions for the child seat and asks for help, experimenter can
show them where they are.
If subject asks experimenter questions, say:
I’m not allowed to help you, but you can find information about that in the manuals for
the child seat and the vehicle.
If subject asks the experimenter to assist with a particular task, say:
I’m sorry, I’m not allowed to help you. Just do your best without hurting yourself or
getting too frustrated.
If subject says I can’t do this, state:
OK, why don’t you try putting it in using (LATCH/seatbelt) instead.
If subject asks what LATCH is, state:
You can find out about LATCH in the manuals for the child seat and the vehicle.
If subject asks how they did, experimenter is allowed to provide a general assessment
such as:
You did pretty good or You improved between the first and last or There are some areas
that could be improved like tightness of the installation
Here is information about the things we are looking at, and here is information about how
you can get your car seat checked at the UM hospital.
Provide subject with SafetyBeltSafe handout on “Quick Checklist for Safety Seat Misuse”
and flyer for Mott Buckle Up Hotline (fitting station at UM hospital).
75
Testing Script / Protocol
Questions
If subject can’t find instructions for the child seat and asks for help, experimenter can
show them where they are.
If subject asks experimenter questions, say:
I’m not allowed to help you, but you can find information about that in the manuals for
the child seat and the vehicle.
If subject asks the experimenter to assist with a particular task, say:
I’m sorry, I’m not allowed to help you. Just do your best without hurting yourself or
getting too frustrated.
If subject says I can’t do this, state:
OK, why don’t you try putting it in using (LATCH/seatbelt) instead.
If subject asks what LATCH is, state:
You can find out about LATCH in the manuals for the child seat and the vehicle.
If subject asks how they did, experimenter is allowed to provide a general assessment
such as:
You did pretty good or You improved between the first and last or There are some areas
that could be improved like tightness of the installation
Here is information about the things we are looking at, and here is information about how
you can get your car seat checked at the UM hospital.
Provide subject with SafetyBeltSafe handout on “Quick Checklist for Safety Seat Misuse”
and flyer for Mott Buckle Up Hotline (fitting station at UM hospital).
76
Appendix D: Subject Evaluation Form – Task 2
Appendix D
Subject Evaluation Forms
77
Appendix D: Subject Evaluation Form – Task 2
Subject ID:
Vehicle: A B C D
Installation number: 1 2 3 4
CRS: C6 C10
Configuration: RF FF Method: L SB Both
Check one answer for each question
Do you agree with these
statements?
Strongly
Disagree
Disagree
I attached the child seat to
the vehicle correctly.
The vehicle manual
matched the child seat
manual.
What I did today is similar
to what I would do at home
to install a child seat.
The vehicle headrest made
it hard to install.
The stiffness of the vehicle
seat made it hard to install.
The shape of the vehicle
seat made it hard to install.
The seatbelt buckles got in
the way of using LATCH.
78
Neutral
Agree
Strongly
Agree
NA
Appendix D: Subject Evaluation Form – Task 2
For seatbelt installations
How hard or easy was it
to:
Very
Hard
Hard
Understand the labels on the
child seat
Understand the instruction
manual about installing the
child seat
Understand the vehicle
instruction manual about
installing the child seat
Figure out how to lock the
seat belt
Figure out where to route the
vehicle belt
Tighten the vehicle seat belt
Figure out what angle the
child seat should be
Adjust the angle of the child
seat
Use the lock-offs on the
child seat that pinch the
vehicle belt
Find the tether anchorage in
the vehicle
Attach the tether strap on the
top of the child seat to the
vehicle
Tighten the tether strap on
the top of the child seat
Store the LATCH belt
Store the top tether (if not
used)
79
Neutral
Easy
Very
Easy
NA
Appendix D: Subject Evaluation Form – Task 2
For LATCH installations
How hard or easy was it
to:
Very
Hard
Hard
Understand the labels on the
child seat
Understand the instruction
manual about installing the
child seat
Understand the vehicle
instruction manual about
installing the child seat
Find the lower anchorages in
the vehicle
Find the tether anchorage in
the vehicle
Figure out where to route the
LATCH belt
Attach the LATCH belt
connectors to the lower
anchorages
Tighten the LATCH belt
Figure out what angle the
child seat should be
Adjust the angle of the child
seat
Attach the tether strap on the
top of the child seat to the
vehicle
Tighten the tether strap on
the top of the child seat
Store the tether (if not used)
80
Neutral
Easy
Very
Easy
NA
Appendix D: Subject Evaluation Form – Task 2
Put an S in all the positions where you could install a child seat using the seatbelt.
Put an L in all the positions where you could install a child seat using LATCH.
Put a T in all the positions where you can attach a top tether.
front of vehicle
Driver
81
Appendix D: Subject Evaluation Form – Task 2
Subject ID:
Date:
Which method did you like best for installing child seats forward-facing (circle one)
LATCH
Seat belt
When thinking about installing child seats, please give each vehicle a rating about how
much you liked it. 1 is worst, 10 is best.
Order Name of Vehicle 1 2 3 4 5 6 7 8 9 10
1
2
3
4
Do you have any suggestions or comments on the vehicles?
82
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising